A new genus of Tetracampidae (Hymenoptera) from South Africa

Afrocampe gen. nov. is described for its only species, A. prinslooi gen. et sp. nov., from Eastern and Western Cape, South Africa. This new genus is recognized as a member of the subfamily Tetracampinae Förster, 1856 of Tetracampidae Förster, 1856 based on the possession of 5-segmented tarsi in females and 4-segmented tarsi in males, the short straight calcar, the mesoscutum with distinct notauli, the mesoscutellum with two pairs of setae, the reduced mesopleural suture and the short stigmal and long postmarginal veins of the fore wing. Afrocampe gen. nov. is characterized by a large mesosoma, a non-convex fi rst gastral tergite, an evenly acute calcar, a 5-segmented antennal funicle, a head lacking occipital carina and facial grooves and a long fore wing with distinctly delimited speculum, a bare admarginal area with a distinct admarginal row of setae on the underside and with 3 setal tracks (hair rows) radiating from the apex of the stigmal vein. The combination of these characters suggests a special status of the new genus within the subfamily Tetracampinae. Moreover, Afrocampe gen. nov. bears some resemblance to the Australian tetracampine genus Niticampe Bouček, 1988. The position of the latter in Tetracampinae, as well as habitus features of the former, are discussed.


Introduction
Tetracampidae Förster, 1856 is a relatively small family of Chalcidoidea Latreille, 1817 (Hymenoptera Linnaeus, 1758), with a complicated taxonomic history (Bouček 1958). The presently classifi ed taxa in this family had previously been assigned to other families until Bouček (1958) proposed the current composition of Tetracampidae. The family was often reported as 'intermediate' between Pteromalidae Dalman, 1820 and Eulophidae Westwood, 1829 (Yoshimoto 1975;Bouček 1988), its contemporary diagnosis includes the possession of a short protibial spur (calcar), the antenna with at least 5 funicular segments, the mesoscutum with complete notauli and a 5-segmented tarsi in females (Noyes 2018). However, the results of recent research suggest that Tetracampidae is a polyphyletic or paraphyletic group, since its subfamilies do not form a solid clade either in trees based on molecular (Munro et al. 2011) or on combined molecular and morphological data (Heraty et al. 2013). Gumovsky (2016)

R e s e a r c h a r t i c l e
urn:lsid:zoobank.org:pub:0EEE6A58-B26A-4592-984D-3C2264A2AB89 reviewed the genera of Tetracampinae and reported the subfamily as consisting of 8 genera, 3 of which occur in the Afrotropical biogeographic realm (sensu Olson et al. 2001).
An enigmatic group of chalcidoids ( Fig. 1) was discovered by the author while working with the collections of the Iziko South African Museum (Cape Town, South Africa, SAMC) and the South African National Collection of Insects, ARC-Plant Protection Research Institute (Pretoria, South Africa, SANC). Although these chalcidoids were habitually more similar to the eulophids of the subfamily Tetrastichinae ( Fig. 1) rather than to any known tetracampids, the possession of the characters mentioned above unambiguously suggested their membership in Tetracampidae. The description of the genus and a discussion of its position in Chalcidoidea are provided below.

Material and methods
Color images for plates were acquired through the Leica LAS 4.4 multi-stacking imaging system in SAMC.  Gibson (1997) and Heraty et al. (2013).
The following abbreviations are used: The term 'mesosoma' means the thorax and propodeum together, and 'metasoma' is a complex of petiole and gaster (Gibson 1997;Heraty et al. 2013).

Etymology
The specifi c epithet is a combination of Afro-(refl ecting the Afrotropical distribution of the genus) and -campe (from Tetracampe). Gender feminine.

Description
Body with ovoid head ( hypopygium not reaching apex of metasoma, mucro short.

Comparative notes
The new genus is very different from all known chalcidoids, but may be assigned to the subfamily Tetracampinae of the family Tetracampidae as judged chiefl y from the combination of the 5-segmented tarsi in females, the 4-segmented tarsi of males and the short straight calcar (see Discussion). Afrocampe gen. nov. has some characters in common with Niticampe Bouček, 1988: the reduced occipital carina, the missing frontal grooves, the evenly acute calcar, the fl at Gt1, the poorly setose base of the fore wing and the distinct forewing speculum. However, Afrocampe gen. nov. differs from Niticampe in general habitus, with a large convex mesosoma (slender in Niticampe), the 5-segmented antennal funicle (6-segmented in Niticampe) and by the setation of the forewing disc (see Discussion).

Biology
Not known.

Etymology
The species is named after Gerhard Prinsloo, a famous expert on South African Chalcidoidea, who collected part of the type series and also fi rst noticed and sorted out the SANC specimens as unusual tetracampids.
HEAD. In dorsal view about 2.0 × as long as broad, vertex smoothly transits to occiput, so occipital margin not traceable (Fig. 2D). POL about 2.0 × OOL, OOL about 2.3 × MDO. Head in frontal view 1.2 × as wide as high; eye approximately 2.3 × as high as broad; ventral margin of torulus situated slightly above virtual line connecting lower margins of eye orbits (Fig. 2B). Minimum distance between inner orbits about 0.6 × width of head. Face smooth, width of scrobal depression about 0.3 × that of face (Fig. 2B). Mouth opening about 1.3 × as wide as malar space (malar space about 0.7-0.8 × as long as mouth opening, Fig. 3F). Inner margins of eyes slightly diverging. Genae nearly straight. Mandibles oriented downwards, bidentate (Fig. 3F). Malar suture sulcate, complete. Antenna (Fig. 3E) with scape about 4.8 × as long as wide, pedicel 1.7 × as long as broad. F1 1.5-1.6 × as long as broad, F2 1.2 × as long as broad, F3 1.3 × as long as broad, F4 and F5 slightly longer than wide, club slightly more than 2.0 × as long as broad, club 3.0 × as long as F5, its segments separated by sutures, terminal spine very short.
LEG. Front leg with a short calcar, about as long as width of its tibia, evenly acute and nearly straight; midtibial spur about 2.0 × as long as width of its tibia; hindtibial spur 1.2 × as long as width of its tibia.

Biology
Unknown.

Distribution
South Africa: Eastern and Western Cape.

Discussion
Afrocampe gen. nov. is considered here a member of the family Tetracampidae based on the possession of the characters traditionally used to distinguish the family (Bouček 1958(Bouček , 1988Gibson et al. 1997;Gumovsky 2016): the short, straight calcar, the 5-segmented tarsus of the female and the mesoscutellum with two pairs of bristles. Heraty et al. (2013) and Gumovsky (2016) suggested that the other extant and fossil groups assigned to this family are likely not related to the nominal subfamily Tetracampinae. Heraty et al. (2013) reported Tetracampinae as monophyletic based on 6 synapomorphies. However, none of these character states are present in Afrocampe gen. nov. Indeed, the male antenna possesses 5 funicular segments in the new genus (not 6, as in most Tetracampinae), the head lacks the upper ocular and transfacial sulci (present in most Tetracampinae), the clypeus is not infl ected into the oral cavity (more or less infl ected in most Tetracampinae), and the forewing speculum is distinctly delimited (not delimited in most Tetracampinae). The presence of an isolated large bristle on each posterolateral corner of the pronotum (as in other tetracampines) is questionable in Afrocampe gen. nov., because in A. prinslooi gen. et sp. nov. the setae in that area are hardly distinguishable from the other dorsal pronotal setae. Gumovsky (2016) listed a number of characters that are diagnostic for Tetracampinae, 5 of which are also present in Afrocampe gen. nov.: 1) the short, straight calcar, 2) the 5-segmented female tarsus, 3) the mesoscutellum with two pairs of setae, 4) the mesopleural suture scarcely evident and 5) the short STV and long PMV. However, the new genus does not possess 5 of the other diagnostic characters for the subfamily, namely: 1) the complete occipital carina, 2) the distinct V-shaped frontal sulcus, 3) the swollen Gt1, 4) the densely setose base of fore wing and 5) the 6-segmented antennal funicle. These characters, apart from the latter, are also absent in the Australian genus Niticampe, which may suggest a relationship with Afrocampe gen. nov. Furthermore, both Niticampe and Afrocampe gen. nov. possess an evenly acute calcar (foretibial spur), as in representatives of the subfamily Mongolocampinae Sugonjaev, 1971 of Tetracampidae, whereas the calcar is bifi d in other tetracampines. However, Niticampe has a 6-segmented antennal funicle, which suggests its affi nity with Tetracampinae, whereas the funicle is distinctly 5-segmented in Afrocampe gen. nov., unlike most other known tetracampines. The 5-segmented funicle is also shared with Mongolocampinae. Also, the fore wing of Afrocampe gen. nov. bears a distinct admarginal hair row on the underside and hair rows which radiate from STV (Fig. 4C, as in the genus Euderus Haliday, 1844 of Eulophidae). Such a forewing setation is not represented in any of the currently described tetracampines. The habitus of Afrocampe gen. nov. is very unusual for Tetracampidae. The robust mesosoma and long fore wing (Figs 2A, C, 4A) differentiate A. prinslooi gen. et sp. nov. from other tetracampids. The combination of a yellow and metallic green colouration is also uncommon among tetracampids, apart from the members of Mongolocampinae, with which Afrocampe gen. nov. also shares the evenly acute calcar and an antenna with one anellus and 5-segmented funicle, but differs in the structure of the mesosoma and forewing venation.
All the above-mentioned considerations make a subfamily attribution problematic for Afrocampe gen. nov. However, it is regarded here as a representative of Tetracampinae, based on the following considerations. First, Afrocampe gen. nov. shares dimorphism in its tarsomere count (5-segmented tarsi of females, 4-segmented of males) with other Tetracampinae. The tarsomere dimorphism does not occur in other Tetracampidae, and may also be considered a putative synapomorphy of the subfamily, even though it was not treated prominently enough in recent phylogenetic studies (Heraty et al. 2013). Second, the 5-segmented funicle is very rare, but not entirely unique for Tetracampinae: it was recorded for Foersterella scaposa Bouček, 1988(Bouček 1988. Third, Afrocampe gen. nov. shares many derived characters with the genus Niticampe, which is, however, characterized by a 6-segmented antennal funicle, similar to most other tetracampines. Both genera are also characterized European Journal of Taxonomy 447: 1-13 (2018) by an evenly acute, non-bifi d calcar and may represent a derived cluster within Tetracampinae. The Cape fl oristic region of South Africa is characterized by a high level of regional and local endemism (Cowling & Holmes 1992;Cowling et al. 1996). The type series of Afrocampe gen. nov. is restricted to this region: despite rigorous and thorough surveys of museum collections and numerous samples collected by the author in Africa, no additional specimens of A. prinslooi gen. et sp. nov. were recorded outside the Eastern and Western Cape Provinces. Such a restricted distribution suggests that Afrocampe gen. nov. is endemic to this region.
When observed among other chalcidoids in a sample (Fig. 1), a specimen of A. prinslooi gen. et sp. nov. is habitually reminiscent of many representatives of the subfamily Tetrastichinae of Eulophidae. The most remarkable similarity in color and body shape is reminiscent of the species of the genera Aprostocetus Westwood, 1833 and Sigmophora Rondani, 1867, which are often associated with galls, in their turn, either as parasitoids or as gall-formers (Noyes 2018). Interestingly, the representatives of the abovementioned subfamily Mongolocampinae, with which Afrocampe gen. nov. shares the structure of the calcar and antennae, are also reported to be either gall formers or gall-former parasitoids (Sugonjaev & Voinovich 2003). Though habitual appearance is not a reliable signal of host association, one can hypothesize that A. prinslooi gen. et sp. nov. is also a gall associate, as in mongolocampines or some tetrastichines. Therefore, a survey for the host of Afrocampe gen. nov. should target the gall formers on endemic plants of the Cape fl oristic region.