Revision of the genus Stenogephyra Lyneborg (Diptera: Therevidae: Phycinae) Author Webb, Donald W. Author Hauser, Martin text Zootaxa 2011 2837 67 85 journal article 47424 10.5281/zenodo.201693 0c0c5871-deb1-4964-b50c-4102bebdbbfd 1175-5326 201693 Stenogephyra Lyneborg Stenogephyra Lyneborg 1987 : 470 (diag.). Type species Stenogephyra torrida Lyneborg by original designation. Diagnosis . Individuals are some of the smallest sized stiletto flies ( 3.2–6.1 mm ), overall black with silver-white facial markings and blue eyes. Antenna twice as long as head, composed of two flagellomeres, first one very long, second one small, quadrate without apical spine. Eyes dichoptic in both sexes. Maxillary palpus one segmented. Thorax black with grey pattern, abdomen black, without markings. Costa ending at M 1 . Femora without stiff setae. Male genitalia without hypandrium, ventral apodeme Y-shaped, gonocoxites not fused, tergite eight (T 8 ) nearly square and apical corners have spiracles embedded. Female genitalia with well developed A 1 acanthophorite spines only. Redescription. Adult. Small sized flies ( Fig. 1 ), male body length 3.2–4.8 , 3.9 mm (n= 40 ), female 3.6–6.1 , 4.9 mm (n= 40 ). Head . Ocellar tubercle dark reddish-brown to black, glossy; setae dark brown to black, elongate. Eyes widely separated, greater than width of ocellar tubercle, ommatidia of equal size. Face strongly protruding with black, glossy callus ventral to antennal bases. Frons glossy with variable species-specific pattern of silver pubescence ( Figs. 2, 4, 6, 8, 10, 12, 14 ). Antenna about two times as long as head length, male antenna slightly longer than female antenna; scape cylindrical, longer than wide, longer than pedicel, subequal to width of flagellum, setae absent on medial surface, macrosetae dark brown; pedicel globular, wider than long, setae absent on medial surface; flagellum with two flagellomeres, first flagellomere cylindrical, elongate, second flagellomere short, quadrate or tapered to apex, apical spine not discernible, flagellum over five times longer than wide, over six times length of scape. Parafacial setose. Maxillary palpus one-segmented, cylindrical, apex rounded. Postocular macrosetae in single row or narrow band. Occiput concave, macrosetae absent. Thorax . Macrosetae 2–3 np, 1 sa, 1 pa, 0–1 dc, 1 sc. Postpronotal lobe concolorous with mesonotum. Setae present on cervical sclerite, propleuron, anepisternum, katepisternum, laterotergite, and scutellum; absent on prosternum, proepimeron, anepimeron, meron, metanepisternum, and metakatepisternum; cervical sclerite lacking macrosetae. Wing. Membrane hyaline. Veins brown. Pterostigma brown. Costa ending at M 1 . R 1 lacking setulae. Cell r 4 + 5 narrow, encloses apex of wing; M 1 and M 2 variable in their origin from apex of discal cell; M 2 may or may not reach wing margin; cells m 3 and cup closed, petiolate; anal angle broadly rounded; alula rounded. Legs. Coxae yellow to dark reddish brown, pubescence silver; midcoxa lacking setae on posterior half; hindcoxa with papillate anterior knob; apical macrosetae dark reddish brown, hind coxa lacking posterolateral macroseta. Femora lacking lanceolate, appressed setae on dorsal surface. Abdomen . Tergites black, glossy; setae short. Male Terminalia . Tergite 8 ( Fig. 16 ) more than 3 times broader than long, with setae only at the apical corners and sternite 8 nearly quadrate, less than 2 times as broad as long, with setae on the whole apical margin ( Figs. 17 , 22, 28, 32, 36 ). Epandrium ( Fig. 16 ) quadrate, median length subequal to width, posterolateral margins short, broadly pointed. Cerci ( Fig. 16 ) wedge-shaped, not fused basally. Hypoproct ( Fig. 16 ) ending distal to posterolateral margin of epandrium. Hypandrium absent. Gonocoxite ventral view ( Figs. 17 , 22, 28, 32, 36 ) separated medially, inner gonocoxal process absent; dorsal view (Fig. 23 ) gonocoxal apodeme narrow, not extending anteriorly beyond anterior margin of gonocoxite. Gonostylus ( Figs. 17 , 22, 28, 32, 36 ), narrow, flattened dorsoventrally, sinuate. Aedeagus with dorsal apodeme (Fig. 23 ) forming broad gonocoxal bridge; distiphallus dorsal (Fig. 23 ), tapered posteriorly; ejaculatory apodeme lateral view ( Figs. 19 , 25, 29, 33, 37 ) broadly enlarged anteriorly, extending beyond anterior margin of dorsal apodeme. Female Terminalia (Fig. 40 ). Tergite 8 is nearly square with a pair of spiracles in the anterior corners. Sternite 8 ( Figs. 20 , 26, 30, 34, 38 ) generally oval, sides rounded, anterior margin variable; aedeagal guide sclerotized. Tergite 10 (acanthophorite) with one set of spines, which are well developed. Sternite 10 rectangular nearly longer than broad, with posterior rounded emargination. Furca ( Figs. 21 , 27, 31, 35, 39 ) closed anteriorly and posteriorly, with two unsclerotized areas, the anterior one is often roundish and smaller. Common spermathecal duct shorter than length of furca, entering in the anterior furcal opening. Three spermathecae. Spermathecal sac absent. Immature stages . From one species ( S. torrida ) we were able to rear several specimens from the last larval stage to adult. The larvae were sieved from loose sand and kept individually in small containers with sand. The larvae were provided 2–3 times weekly with a flower beetle larvae ( Tribolium sp.) and a drop of water. The time between pupation and eclosion was on average 7 days ( 6–9 , n= 10 ). Pupa (Figs. 41–43 ). On the top of the head is a pair of bumps, which we describe here for the first time with the name occipital sheath. The antennal sheath is long and in a 45 ˚ angle with the body axis. The labral sheath has at its apex a small incision. The alar process is absent. The abdominal segments have at least 1 pair of abdominal spines on segments 2 and 3 , while segments 4–7 often have an additional pair of shorter spines. The female pupa has no spines on segment 8 , while male pupa possess a pair of short ventral spines. This sexual dimorphism could be found in Phycinae and Therevinae ( Hauser & Irwin 2003 ). The other sexual dimorphism is manifested in the length of the antennal sheath, which is longer in males of Stenogephyra than in females. In females the tip of the antennal sheath is at the height of the base of the labral sheath, while in males the tip of the antennal sheath is at the height of the tip of the labral sheath. The pupa exhibits several characteristics typical for a Phycinae : the antennal sheath is at a 45 ˚ angle to the body axis (in Therevinae the antennal sheath is at a right angle to the body axis); the lateral antennal process is shorter than the antennal sheath pore (in Therevinae the lateral antennal sheath process is significantly longer that the antennal sheath pore); and the alar process is absent (present in Therevinae ). For comparison with a therevine pupa see Hauser & Irwin ( 2003 ) . Species included: Stenogephyra dianeae spec. nov. Stenogephyra janiceae spec. nov. Stenogephyra minuta Lyneborg Stenogephyra namibiensis spec. nov. Stenogephyra parkeri spec. nov. Stenogephyra schlingeri spec. nov. Stenogephyra torrida Lyneborg Comments on Lyneborg’s characters . Lyneborg ( 1987 ) indicated that Stenogephyra and Phycus Walker form a monophyletic group and considered them sister taxa. Below, we discuss the characters (three synapomorphies and four sympleisiomorphies) he originally used to place Stenogephyra together with Phycus , and then present our ideas about the genera related to Stenogephyra and their biogeographical significance. Synapomorphies sensu Lyneborg ( 1987 ) : 1 . "First flagellomere elongate." This character is clearly shared between Phycus and Stenogephyra , but it is also found in the South American phycine genus Ataenogera Kröber ( Hauser and Webb 2007 ) . 2 . "Femora without setae." The femora of all examined Phycus species and most specimens of Stenogephyra do not possess stiff dark setae, which are common in most other Therevidae genera. Although species like S. dianeae spec. nov. have some stiff black or white setae at the apex of the hind femur, it is not clear if those are homologous to the stiff black setae on the hind femora of other Phycinae . The New World genera Ataenogera , Pherocera Cole, Parapherocera Irwin , and Schlingeria Irwin do not possess these setae either. 3 . "Hypandrium vestigial or absent." In all species of Stenogephyra and most species of Phycus , the hypandrium is absent. This character state can also be found in Pherocera , Parapherocera and Schlingeria , but in Ataenogera the hypandrium is well developed. Sympleisiomorphies sensu Lyneborg ( 1987 ) : 1 . "Male eyes broadly separated." Ataenogera and Parapherocera have dichoptic males, while several Phycus (e.g. Phycus angustifrons Lyneborg ) have eyes only separated by the width of the median ocellus. 2 . "Maxillary palpus two-segmented." Among the phycine genera only Phycus has a two-segmented maxillary palpus. In the case of Stenogephyra, Lyneborg was clearly wrong, which might be attributed to the small size of these flies and the difficulty to discern such structures. The maxillary palpus of Stenogephyra is one segmented. 3 . "Prosternum bare." Most phycine genera lack setae in the prosternal depression and this character is sometimes not consistent within a genus. 4 . "Ventral epandrial sclerite (hypoproct) large." This character was not thoroughly investigated by the authors, but is known to be present in several other genera. Stenogephyra has a set of unique characters (synapomorphies) that define it as a monophyletic group: The tergite 8 in females being nearly square with spiracles in each anterior corner, while all other Phycinae have the tergite 8 broader than long and without spiracles. The tergite 10 (acanthophorite) in females has one set of spines, which are exceptionally well developed for a member of Phycinae ; most Phycinae have much shorter and less pronounced spines. The sternite 10 in females is rectangular and nearly longer than broad, with posterior rounded emargination. While all examined Phycinae also have a more or less distinct posterior emargination, sternite 10 in all other Phycinae genera is much broader than long. In higher Therevidae the sternite 10 has a medioapical extension, while the Xestomyzinae have an oval shaped sternite 10 with a long and thin anterior extension. Despite the fact that some of Lyneborg's characters which linked Phycus with Stenogephyra do not hold anymore, there are still several characters which both genera share. In a combined morphological and molecular dataset ( Hauser 2005 ), Stenogephyra is the sister–genus to the Nearctic Pherocera –clade ( Pherocera , Parapherocera and Schlingeria ) and together they form the sister-group to Phycus . This result was very surprising, because of the external morphological differences and the geographical distances between the Nearctic Pherocera clade and the South African Stenogephyra . Still, the morphological data show some interesting shared characters, which make a close relationship between these groups more plausible. For example, all Stenogephyra , Pherocera –clade species and Phycus have a Y-shaped ventral apodeme, lack a hypandrium, do not have the gonocoxites fused, and lack setae on the femora. Parapherocera and Stenogephyra have holoptic males with a protruding face. Tergite eight (T 8 ) in Stenogephyra is nearly square and the apical corners have the spiracles imbedded, while in most other examined Phycinae , T 8 is much wider than long and without spiracles. The notable exceptions are Pherocera , Parapherocera and Schlingeria , which have T 8 more squared. Phycus is distributed throughout Africa, the Mediterranean region, and east through Asia and into western North America . It is very likely that the genus originated in Africa, where most species are found, and spread to the New World through Asia and never colonized South America . We could hypothesize a similar distribution for the common ancestor of Stenogephyra and the Pherocera -clade, and suggest extinction in most parts of the distribution. Also, this dispersal event should be older than the spread of Phycus , as in the case of Stenogephyra and the Pherocera -clade, the morphological differences are much greater and on a generic level, while the North American species of Phycus are still clearly congeneric with their African and Asian counterparts. Hopefully, a more detailed phylogenetic analysis will shed some light in future on the relationships and the biogeographical scenarios.