A review of the assassin-fly genus Anypodetus Hermann, 1907 with the description of a new species (Insecta, Diptera, Asilidae)

The genus Anypodetus Hermann, 1907 (Diptera, Asilidae, Laphriinae) is reviewed. Currently, eight species are recognized from Botswana, Mozambique, Namibia, South Africa, Zimbabwe, and southern-most Zambia, i.e., Anypodetus arachnoides Oldroyd, 1974 widespread, Anypodetus fasciatus Hermann, 1907 widespread, Anypodetus fascipennis Engel, 1924 widespread, Anypodetus leucothrix Londt, 2000 restricted to southern Namibia and south-western South Africa, Anypodetus macroceros Londt, 2000 restricted to west-central Namibia, Anypodetus nigrifacies Ricardo, 1925 restricted to eastern-most South Africa and southern Mozambique, Anypodetus phalaros Londt, 2000 Namibia and South Africa, and Anypodetus unicolor Oldroyd, 1974 Namibia, eastern South Africa, adjacent Mozambique, and southern Zimbabwe. One new species, Anypodetus londti sp. nov. from Mozambique and Zimbabwe, is described for a total of nine species in the genus. Study of the secondary type specimens of A. unicolor from Namibia revealed that these specimens do not represent this species, reducing the number of species recorded from Namibia to six. Anypodetus leucothrix is recorded with several additional collecting events in central and northern Namibia extending its range significantly. Distribution, biology, occurrence in biodiversity hotspots sensu Conservation International, and seasonal imago flight activity are discussed. Diagnoses, photographs, specimen occurrence data, and an identification key to species are provided with the new species described in detail. The sexual dimorphism in the development of the mystax and wing vein variation in regard to the alignment of M2 and M3 are discussed and illustrated. African Invertebrates 64(2): 165–206 (2023) doi: 10.3897/AfrInvertebr.64.104283 https://africaninvertebrates.pensoft.net Copyright Torsten Dikow & Meliah Dubus. This is an open access article distributed under the terms of the CC0 Public Domain Dedication. RESEARCH ARTICLE Torsten Dikow & Meliah Dubus / African Invertebrates 64(2): 165–206 (2023) 166


Introduction
Anypodetus Hermann, 1907 is a morphologically unique genus of assassin flies. It is one of two Afrotropical Laphriinae genera without pulvilli (Londt and Dikow 2017a, Figs 1, 12, 16) -the other one being Prytanomyia Özdikmen, 2006 (Londt and Dikow 2017b). Furthermore, the majority of species exhibit a unique arrangement of wing veins in that the proximal portion of vein M 2 and the distal portion of vein M 3 form a line from anterior to posterior with the crossvein m-m being absent or highly reduced (Figs 13,19,23,31,36,47,49). This venation is known from genera of Atomosiini (Laphriinae, Hermann 1912;Hull 1962;Dikow 2009) including the Afrotropical genera Dichaetothyrea Meijere, 1914, Goneccalypsis Hermann, 1912, and Loewinella Hermann, 1912 reviewed by Londt (1982) as well as Orthogonis Hermann, 1914 (Londt andDikow 2017a). Anypodetus fascipennis Engel, 1924 is the exception as it is the only species of this genus with a distinct crossvein m-m (Figs 25,29) but it is important to note that there appears to be intraspecific variation in the arrangement of these veins (see Discussion). Another interesting morphological feature is that several species exhibit distinct mystax morphology in females and males. In Asilidae, the sexes are usually only distinguishable morphologically by examining the tip of abdomen and features such as the mystax cannot be used to determine the sex of a specimen with a broken abdomen.
Anypodetus was reviewed by Londt (2000) who recognized eight species, which are restricted to southern Africa with a single specimen record from southern Zambia to date. These species have been collected throughout southern Africa (Figs 1-11). The highest species diversity is found in Namibia with seven recorded species (Londt 2000, Figs 51, 52, but see below).
This study was instigated by the discovery and collection of unique flies belonging to Anypodetus in the Namib Desert in west-central Namibia by the senior author that did not key out immediately to one of the known eight species in the identification key published by Londt (2000). Further study of these specimens revealed that they belong to a described species, A. leucothrix Londt, 2000 (Figs 30-35), but that they exhibit variation in the wing venation not originally described for this species. In addition, an undescribed species was discovered among natural history museum collections and is described below. This study was impacted by the Sars-CoV-2 pandemic in that not all specimens were available as a loan from museums. Even in the absence of seeing every single specimen, the identifications by Londt (2000) are regarded as accurate and have been included in the material examined adding their unique specimen identifier (NMSA-DIP-XXXXX) now established at the NMSA for future reference. Nonetheless, this review includes many specimens not studied by Jason Londt in 2000 from the NMBZ, SMNS, and USNM that greatly expand our understanding of the distribution of Anypodetus species (see also Fig. 11).  (1) and T. Dikow (2)(3)(4).
The taxonomic history of Anypodetus can be summarized as follows: Hermann (1907) described the genus Anypodetus with its type species Anypodetus fasciatus Hermann, 1907 from Lichtenburg in the North-West province of South Africa. Engel (1924) described Anypodetus fascipennis Engel, 1924 and Anypodetus semirufus Engel, 1924 both from Willowmore, Eastern Cape province of South Africa. Ricardo (1925) described Anypodetus maculipennis Ricardo, 1925 from Sawmills and Bulawayo, Zimbabwe and Anypodetus nigrifacies Ricardo, 1925 from Lourenço-Marqués (= Maputo), Mozambique. Engel (1929) highlighted the fact that J. Brauns, the collector of the types series of both A. fascipennis and A. semirufus at Willowmore, suggested that A. semirufus is a variety of A. fascipennis with which Engel agrees, therefore synonymyzing the two species. He furthermore synonymized A. maculipennis with A. semirufus (an unusual action since semirufus was just made a junior synonym) and A. nigrifacies with A. fascipennis (an action no other authors have accepted).  Engel and Cuthbertson (1934) Engel's (1929) and Engel and Cuthbertson's (1934) earlier actions, recognizing six valid species. Oldroyd (1980) Londt (1998) reported on the species of Asilidae recorded from the southernmost Nama Karoo in and around the town of Willowmore by J. Brauns in the NMSA collection and listed three Anypodetus species identified by F. Hermann: Figure 11. Map of southern Africa with elevational relief and biodiversity hotspots (sensu Conservation International in gray) and distribution of Anypodetus specimens studied by Londt (2000)  At the commencement of this study, Anypodetus was, therefore, known from eight species: A. arachnoides, A. fasciatus, A. fascipennis, A. leucothrix, A. macroceros, A. nigrifacies, A. phalaros, and A. unicolor.

Materials and methods
Morphological features were examined using Zeiss SteREO Discovery.V8 and V12 stereo microscopes. Wing length is measured from the tegula to the distal tip of the wing.

Species description
The species description is based on composites of all specimens and not exclusively on the holotype and is compiled from a character matrix of 230 features assembled with Lucid Builder (version 4.0.10) and eventually exported as natural-language descriptions. The species description includes therefore features that might not vary within Anypodetus but represents a comprehensive morphological description to al-low future species discoveries and comparisons to other Asilidae genera. The species description has been deposited in the Zenodo data depository and can be accessed in XML-format following the SDD (Structure of Descriptive Data) standard. All taxon names have been registered in ZooBank (Pyle and Michel 2008). If available, permanent URLs or Digital Object Identifiers (DOIs) to the original species descriptions on the Biodiversity Heritage Library (BHL, www.biodiversitylibrary.org) or other online sources are provided. The species record for each species at the Global Biodiversity Information Facility (GBIF, www.gbif.org) provides a summary of occurrence data, images, or taxonomic treatments from natural history collections. The species descriptions from Londt (2000) have been marked up in TaxonX XML language (Catapano 2010) and uploaded to the Plazi TreatmentBank from where they are accessible in human-and machine-readable formats and a permanent URL is provided under each respective species.

Specimen occurrence data
The following data on species occurrences are given (where available): country, state/ province, county, locality, geographic co-ordinates, elevation (in meters), date of collection, time of day at collection (if available), habitat information, sampling protocol (if other than hand netting), collector, catalog number (a unique specimen identifier and any other identifying number), depository (institution code), number of specimens, sex, life stage, and any other previous identifications. Each specimen is listed with a unique specimen identifier (either an institutional catalog number or an AAM-XXXXXX number used by the senior author) that will allow the re-investigation as well as provide a unique Life Science Identifier (LSID). The occurrence of all species is illustrated in distribution maps plotted with SimpleMappr (Shorthouse 2010) with all of those localities for which co-ordinates are available or could be gathered from online gazetteers or Google Earth. Type localities are plotted with a square symbol, other specimens are plotted with a circular symbol, and iNaturalist observations are plotted with a star symbol. The distribution maps include Biodiversity Hotspots sensu Conservation International (Mittermeier et al. 1998;Myers et al. 2000;Mittermeier et al. 2005).

Photographs and illustrations
Whole habitus photographs of pinned USNM specimens were taken with a GIGAmacro Magnify 2 system, a Canon EOS D5 Mark IV full-frame DSLR, a Canon MP-E 65 mm f2.8 macro-lens, and illuminated by a Canon MR-14EX II Macro Ring Lite. Individual RAW-format images were stacked using HeliconFocus Pro (version 8.+) and exported in Adobe DNG-format. Photographs of Smithsonian USNM specimens are in the public domain with a Creative Commons license CC0 and can be downloaded in full resolution from the USNM data portal (http://collections.nmnh.si.edu/search/ ento/) or the Smithsonian Open Access Portal (https://www.si.edu/openaccess).

Keys
The online, interactive dichotomous key has been built with Lucid Builder (version 4.0.10) and can be accessed on Lucidcentral and the senior author's research web-site. It has also been archived in the Structure of Descriptive Data (SDD) standard at Zenodo.

Institutions providing specimens
Institutions providing specimens are listed below, together with the abbreviations used in the text when citing depositories (  Diagnosis. The genus can be delineated by the absence of pulvilli, very long macrosetae on the scape that often reach the tip of the antennae, the sexual dimorphism in mystax setae coloration and arrangement in some species, the absence of macrosetae on the supero-posterior anepisternum, the wing cell r 5 open and cell m 3 closed, the small size with a wing length of 4.8-8.5 mm, and the restricted distribution to southern Africa (single record from southern Zambia).
Distribution, biodiversity hotspots, seasonal imago flight activity, and biology. Known from throughout southern Africa with a single locality in southern Zambia Table 3. Seasonal imago flight activity of Anypodetus species through number of specimens collected and unique collecting events in each month (data given as # specimens/# collecting events). Months abbreviated starting with July. * = additional iNaturalist observation.    (Figs 11,51,52). A relatively commonly observed and collected genus known from 378 specimens in museum collections from 131 collecting events between 1906-2019 and nine observations at iNaturalist ( Table 2). Three species of Anypodetus occur in the Maputaland-Pondoland-Albany biodiversity hotspot, but none are endemic to this hotspot. Adult flies are active from spring to late summer (September-March) with a single record for August (Table 3). With the exception of a few isolated prey records nothing is known of the biology. Plazi TreatmentBank (Londt 2000):

Anypodetus arachnoides
Diagnosis. The species is distinguished from congeners by the entirely orange postpedicel, the short white, tightly packed, dorso-ventrally flattened macrosetae in the male mystax, and the entirely black mystax in females restricted to the lower facial half. Distribution, biodiversity hotspots, seasonal imago flight activity, and biology. Known from throughout southern Africa with a single locality in southern Zambia but not recorded from Mozambique to date (Fig. 51). A commonly observed and collected species known from 67 specimens from 23 collecting events between 1913-2005 and three observations at iNaturalist ( Table 2). The species is not known to occur in any currently recognized biodiversity hotspot. Adult flies are active from spring to late summer (September-March) with a single record for August (Table 3). Londt (2000) reports two prey records: Diptera: Platystomatidae and an alate Formicidae. Hermann, 1907 Figs 1, 2, 18-23, 52 Anypodetus rigidis Oldroyd, 1974 -  Distribution, biodiversity hotspots, seasonal imago flight activity, and biology. Known from throughout southern Africa except Mozambique to date (Fig. 52). A commonly observed and collected species known from 85 specimens from 35 collecting events between 1919-2018 and a single observation at iNaturalist ( Table 2). The species is not known to occur in any currently recognized biodiversity hotspot. Adult flies are active from late spring to late summer (October-March) ( Table 3). Engel and Cuthbertson (1934) report grasshoppers as prey but no predator or prey specimens were studied by them. Specimens collected by the senior author have been recorded on or near sand dunes perching on sand or vertically on low vegetation (see Figs 1, 2). Engel, 1924 Figs 3, 4, 24-29, 52 Anypodetus semirufus Engel, 1924 - GBIF: https://www.gbif.org/species/1664902. Diagnosis. The species is distinguished from congeners by the brown-stained anterior half of the wings with white transverse bands, the lateral frons being strongly developed with 3-4 short, yellow proclinate macrosetae, the circular macrosetae (female yellowish and black, male white) in the mystax, and the proximal portion of vein M 2 and the distal portion of vein M 3 not aligned.

Distribution, biodiversity hotspots, seasonal imago flight activity, and biology.
Known from throughout southern Africa except Mozambique to date (Fig. 52). A commonly observed and collected species known from 121 specimens from 31 collecting events between 1907-2015 and three observations at iNaturalist ( Table 2). The species is not known to occur in any currently recognized biodiversity hotspot. Adult flies are active from late spring to late summer (October-March) ( Table 3). Specimens collected by the senior author have been recorded perching on ground or on low vegetation (see Figs 3,4).
Remarks. Ricardo (1925: 242) described A. maculipennis on 'Type (male) from Saw Mills, S. Rhodesia. Type (female) from Bulawayo, and other males and females from the same locality (Rhodesia Museum).' In doing so she did not designate a holotype so all her listed specimens must be considered syntypes. The NHMUK specimens of A. maculipennis were labeled as Syntypes by J. Chainey in 1984. Londt (2000: 134) designated 'the male as lectotype and the female as paralectotype' but none of the specimens are labeled as such and there are more than two specimens. We hereby designate the male from Sawmills as Lectotype (NHMUK013445832) in order to stabilize the species concept and the other two female specimens from Bulawayo in the NHMUK as Paralectotypes (NHMUK013445833 and NHMUK013445834). Londt (1998) reported on a specimen from Willowmore that was identified by F. Hermann as 'A. varipennis', which represents an unpublished name. The specimen in the NMSA (NMSA-DIP-008988) has been studied and verified to represent A. fascipennis. Distribution, biodiversity hotspots, seasonal imago flight activity, and biology. Known only from south-western South Africa and Namibia (Fig. 51). A rarely observed and collected species known from 33 specimens from 10 collecting events between 1925-2019 and a single observation at iNaturalist ( Table 2). The species is not known to occur in any currently recognized biodiversity hotspot. Adult flies are active in spring (September-November) and late summer (February) ( Table 3). Specimens collected by the senior author have been recorded on or near vegetated dunes perching on sand.

Anypodetus leucothrix
Remarks. The collection of this species by the senior author in the Namib Desert initiated interest in reviewing this genus. At first, the specimens collected could not be readily identified using the key in Londt (2000). In that key, A. leucothrix is keyed primarily through the wing venation being similar to A. fascipennis in which the proximal portion of vein M 2 and the distal portion of vein M 3 are not aligned (see Figs 25,29). The collected specimens, however, showed a complete (or near complete) alignment of M 2 and M 3 (Figs 31, 35). Because of the considerable distance of the Namib Desert locality to the localities of the two paratypes in southern Namibia and the type locality in south-western South Africa (Fig. 51), it was postulated that the collected specimens represent a species new to science. Only a detailed study of the male terminalia comparing them to the illustrations by Londt (2000) and photographs of the holotype in the NMSA supported the conclusion that the Namibian specimens belong to A. leucothrix. Additional specimens collected in 1954 (in SMNS) and 1963 (in USNM) further north in Namibia added to the understanding that there is wing venation variation within A. leucothrix. Londt (2000) stated that this species is somewhat different from all other Anypodetus species in that the thorax is primarily apubescent (all other known species have a pubescent thorax) and the scutellum has long apical scutellar setae (apical scutellar setae ab-sent in other species). In contrast, the basic morphology of the male terminalia is identical to all other known species of Anypodetus (see figures in Londt 2000). Furthermore, the lack of pulvilli, the absence of a macroseta on the supero-posterior anepisternum, and the sexual dimorphism in the mystax coloration and development are features shared among all Anypodetus species recognized here including A. leucothrix.
Etymology. The species is named after Jason G.H. Londt in celebrating his career with the present Festschrift in the year of his 80 th birthday. Jason is, without doubt, the most knowledgeable Afrotropical Asilidae taxonomist, present and past.
Antenna: light brown to brown, lightly gray pubescent; scape approximately as long as pedicel, short black setose dorsally and long black macrosetose ventrally, macrosetae very long, reaching tip of postpedicel; pedicel short black setose ventrally, long black setose laterally; postpedicel medially broadest, long, approximately 2× as long as scape and pedicel combined, asetose; stylus comprised of 1 element, 0.25× as long as postpedicel, asetose; apical seta-like sensory element situated apically in cavity on stylus.
Wing: 4.6-5.6 mm, hyaline, evenly microtrichose; C circumambient (developed around entire wing), anterior wing margin in males straight; R 2+3 distally relatively straight, r 1 closed, R 1 and R 2+3 meet apically and form a stalk vein (petiolate); R 4 terminating anterior to wing apex, distinctly arching anteriorly, stump vein (R 3 ) absent; r 4 open, R 4 and R 5 diverging from each other; R 5 terminating posterior to wing apex; r 5 open; M 1 terminating posterior to wing apex; cell d closed by base of M 2 , m-m absent (or at least highly reduced), M 2 and M 3 aligned in a line from anterior to posterior, r-m situated in center; m 3 closed and petiolate; cua closed and petiolate; alula welldeveloped; microtrichia on posterior wing margin arranged in a single plane.
Distribution, biodiversity hotspots, seasonal imago flight activity, and biology. Known only from two localities in southern Mozambique and north-western Zimbabwe (Fig. 51). A rarely collected species known only from two specimens (both males) from two collecting events in 1938 and 1964 ( Table 2). The species is not known to occur in any currently recognized biodiversity hotspot. Adult flies are active in summer (December and February) ( Table 3). Nothing is known of the biology.
Remarks. While the male terminalia illustrations by Londt (2000) have been useful in identifying A. leucothrix, the species-specific features are difficult to describe sufficiently. Because only two male specimens in less-than-ideal conditions are known for A. londti sp. nov., we decided not to dissect a specimen for illustration in order to preserve entire specimens for future study. Photography of these structures was also not possible because of the orientation of the male terminalia and legs etc. to view all angles properly (Fig. 36). The male terminalia described above provide general features useful to distinguish species of this genus from other Afrotropical Laphriinae. The ventral terminalia aspect with the number of strong macrosetae distally on the gonocoxite is relatively easily viewable in non-dissected specimens. In A. londti sp. nov. there are 3-5 such strong macrosetae present (see figs 11-33 in Londt 2000). Londt, 2000 Figs 39-41, 52 Taxon depository. ZooBank: https://zoobank.org/15C1B571-6E8A-4EE6-95FD-F883C73E619D; GBIF: https://www.gbif.org/species/1664900; Plazi TreatmentBank: https://treatment.plazi.org/id/03B39D2F-F048-9D42-FF7D-FB43D9F14647.

Anypodetus macroceros
Diagnosis. The species is distinguished from congeners by the unique postpedicel shape in which the apical part is narrowing and appearing as an elongate stylus, which itself is developed regularly (see fig. 6 in Londt 2000), and the densely arranged microtrichia on the wing. Distribution, biodiversity hotspots, seasonal imago flight activity, and biology. Known only from the type locality in the central Namib Desert in Namibia (Fig. 52). A rarely collected species known only from a single specimen from one collecting event in 1974 ( Table 2). The species is not known to occur in any currently recognized biodiversity hotspot. Adult flies are active in February in summer (Table 3). Nothing is known of the biology. Ricardo, 1925 Figs 42-45, 52  Remarks. Ricardo (1925: 244) described this species on 'Types (male and female) and another male and female, all from Lorenzo Marques, Portuguese E. Africa (F.D. McMillan), in Brit. Mus Coll.' In doing so she did not designate a holotype so all her listed specimens must be considered syntypes. The three NHMUK specimens of A. nigrifacies were labeled as Syntypes by J. Chainey in 1984. It is unknown where the 4 th specimen is deposited. Londt (2000: 134) designated 'the male as lectotype and the female as paralectotype' but none of the specimens are labeled as such and there are more than two specimens. We hereby designate the male from Lorenzo Marques (= Maputo) as Lectotype (NHMUK013445836) in order to stabilize the species concept and the two female specimens from the same locality as Paralectotypes (NHMUK013445835 and NHMUK013445837). Distribution, biodiversity hotspots, seasonal imago flight activity, and biology. Known only from north-eastern South Africa, southern Botswana, and south-central Namibia (Fig. 52). A rarely observed and collected species known from three specimens from three collecting events between 1972-1975 and a single observation at iNaturalist in 2015 ( Table 2). The species is not known to occur in any currently recognized biodiversity hotspot. Adult flies are active in summer (January and March) ( Table 3).
Diagnosis. The species is distinguished from congeners by the presence of apical scutellar macrosetae, the entirely gray pubescent scutellum, and two black medial macrosetae laterally on abdominal tergites 2-5.

Discussion
Wing venation intraspecific variation Londt (2000) used the alignment of veins M 2 and M 3 in the first couplet in his key to species. As highlighted under A. leucothrix, the alignment of these veins can be variable within a species and may cause misidentifications. Furthermore, it is somewhat difficult to state whether the veins are fully aligned or not -it is clear when they are such as in the A. arachnoides specimen photographed in Fig. 13 but not as straightforward when there is a slight alignment break as in the A. fasciatus specimen photographed in Fig. 23. It is also difficult to discern any particular pattern of this venation variation. It does not appear to be strictly correlated with the distribution as specimens from the same locality may show alignment differences as in the specimens of A. leucothrix collected at the same locality in late September 2017 and 2019 (Figs 31, 35, red arrow). The very distinct non-alignment of M 2 and M 3 in A. fascipennis is unmistakable (Fig. 29) and unique to this species. The newly developed identification key above does not rely on wing venation to avoid any problems in determining Anypodetus specimens.

Mystax development and sexual dimorphism
In Asilidae, the sexes are usually only distinguishable morphologically by examining the tip of the abdomen to locate the female ovipositor or male terminalia. Anypodetus is unique in that respect as the development of the facial setation, the mystax, differs between the sexes in several species. To help identify species and to associate females and males, Table 1 provides short, comparative descriptions of the mystax development in females and males. Of the nine Anypodetus species, seven are known from both sexes (A. londti sp. nov. and A. macroceros are only known from males). Of these, the mystax between females and males differs in four species, i.e., A. arachnoides, A. fasciatus, A. fascipennis, and A. leucothrix. Anypodetus arachnoides females are unique in that the mystax is restricted to the ventral ⅓ of the face (Fig. 14) whereas in all other species, the mystax occupies the entire face. Anypodetus leucothrix males are unique in that the face is entirely apubescent (Fig. 33) whereas it is gray, yellowish, or light brown pubescent in all other species. Also more unusual among Asilidae is that the mystax of Anypodetus is often composed of different macrosetae such as 'regular' circular setae and dorso-ventrally flattened setae, i.e., in A. arachnoides males (Fig. 15), A. londti sp. nov. males (Fig. 38), and A. phalaros females and males (Fig. 48).

Distribution
Species of Anypodetus occur throughout southern Africa but so far the genus has not been recorded from Eswatini or Lesotho (Fig. 11). A notable exception in terms of its distribution is its absence in much of the coastal habitats along the Atlantic and Indian oceans with the exception of eastern-most South Africa and southern-most Mozambique (Fig. 11). Furthermore, Anypodetus has not been collected or observed in the otherwise species-rich Succulent Karoo and Cape Floristic Region biodiversity hotspots (Fig. 11). It is likely that the genus also occurs in at least southern Angola and is more widespread in southern Zambia and Mozambique. Three species, A. arachnoides, A. fasciatus, and A. fascipennis, are more commonly collected (Table 2) and widespread across southern Africa (Figs 51, 52). A. leucothrix (Fig. 51) and A. phalaros (Fig. 52) are somewhat widespread as well, but with many fewer collecting events to date (Table 2). Interestingly, several species have been collected at the same locality, and sometimes even sympatrically. For example, A. fasciatus

Conclusion
Anypodetus is a unique genus of Afrotropical Asilidae restricted to southern Africa. The now nine recognized species are widely distributed in this region and the number of iNaturalist observations might be an indication that species can easily be observed in natural environments. Namibia holds the largest species diversity with six species recorded.