Bittacus chevalieri ( Navás, 1908 )

Londt, Jason G. H., 2001, Bittacus chevalieri (Navás, 1908) (Mecoptera: Bittacidae), newly recorded from Socotra and found to occur in Pakistan and India, African Invertebrates 42, pp. 255-262 : 257-262

Bittacus chevalieri ( Navás, 1908)

( Figs 1–10 )

Haplodictyus chevalieri Navás, 1908: 16 , Navás, 1909: 529.

Bittacus chevalieri ; Esben-Petersen, 1913: 139, 1915: 237, 1921: 136–137; Navás, 1921: 77; Lestage, 1929:

3–4; Fraser, 1953:?; Byers, 1971: 402–3; Londt, 1994: 49.

Bittacus nodosus Rust & Byers, 1976: 32–34 . Syn. n.

Redescription: Based primarily on pinned and alcohol-preserved material from Mali and Socotra in NMSA. Sexes similar unless otherwise stated.

Head ( Fig. 1 ): Brown-yellow, ocelli enclosed within a dark red-brown circular area. Ocelli large, dorsal pair larger than single median ocellus. Face between eyes with a shallow depression surrounded by inwardly directed setae. Antennal scape and pedicel brown-yellow, flagellum brown.

Thorax: Uniform brown-yellow (i.e. no obvious dark red-brown markings), pro- and mesonota slightly darker than other sclerites.All macrosetae black. Legs: Brown-yellow; distal tips of femora, tibiae and tarsomeres dark red-brown. Wings ( Figs 2–3 ): FW 16.5 mm, HW 14.9 mm (average of 5 _5^ from Socotra, measured from humeral crossvein to wing tip, ^ on average only slightly smaller than _); venation dark brown, some distal crossveins weakly demarcated (i.e. poorly sclerotised); membrane uniformly covered with fine microtrichia, giving wings a slight pale brownish appearance. Pterostigma only slightly darker than other parts of membrane. Main features of venation as follows:

• Pcv variable in number (1 or 2) even in a single specimen. In a sample of 5 _5^ from Socotra there are 2 Pcvs in 95% of FWs and 50% of HWs (the more common condition therefore being 2 Pcv).

• Scv, FRs, FM and Cuv almost in line in FW, Scv position being variable from a little beyond an imaginary line drawn through FRs and FM to a little before.

• FRs, FM and Cuv almost in line in HW, Scv position variable, but always somewhat behind an imaginary line drawn through FRs and FM.

• A 1 short, reaching wing margin about halfway between distal tips of Cu 2 and A 2.

• Av absent in all specimens studied (possibly a consequence of having a short A 1). A few specimens possess a small ‘spur vein’ at distal end of A 1 ( Fig. 2 ) which may represent the remnants of an Av.

Abdomen: Pale brownish yellow; proximal terga slightly paler than other sclerites. T3–

8 with narrow dark red-brown posterior margin (seen in cleared specimens only). Sterna,

especially S2–5, darker yellow-brown when compared to corresponding terga. Genitalia:

Similar in colour to preceding segments. _ ( Figs 4–7 ) with distal parts of aedeagus

brown.A few alcohol-preserved males have pheromone glands protruding from between T6–7 and T7–8. ^ ( Figs 8–9 ) with poorly defined (i.e. weakly sclerotised) subgenital plate.

Variation: Material studied suggests a species of remarkably uniform appearance across its range. The slight variation shown between males (especially in the shape of the epandrium) from Socotra and Mali appears to be consistent geographical variation as all males from these places have a generally similar appearance.

Previously recorded material: Afrotropical: CHAD: 1^ holotype, Afrique Central , Kanem [Region of Chad – 15 ° 00'N: 16 ° 00'E], A. Chevalier, 1904 ( MNHN) . DJIBOUTI: 1^, Obok [= Obock – 11 ° 57'47''N: 43 ° 17'26''E], Jousseaune ( MNHN) . MALI: 1^, Dogo [? = Macina – 13 ° 39'N: 05 ° 20'W] (repository not known); 1 _ 1?, Nioro [15 ° 14'N: 09 ° 35'W], Soudan, F. de Zelther ix.1909, ( MNHN) . SENEGAL: 1 _, Matam [15 ° 40'N: 13 ° 15'W] and Bahel [?], Dialla [?], G. Malou, 1908 29.ix.1907 ( MNHN) ; 1 _, Linguère [15 ° 24'N: 15 ° 07'W], A. Descarpentries, T. Leye & A. Villiers, Mission IFAN – Museum, ix.1967 ( MNHN) . UNKNOWN: 1^, Africa ( NHMW) . Oriental: INDIA: 1 _ 1? paratypes ( nodosus ), Deesa [24 ° 15'N: 72 ° 10'E], 8.i.1935, G. G. Nurse ( BMNH) . PAKISTAN: 1 _ holotype ( nodosus ), Karachi [24 ° 52'N: 67 ° 03'E], 8.x.1959, R. I. Sailer

(SEMC). New records: CHAD: 1 _, N’Gouri [13 ° 38'N: 15 ° 22'E], Dist. de Kanem, P. Renaud ,

viii.1958 ( MRAC) ; MALI: 14 _2^1?, Mourdiah [14 ° 28'N: 07 ° 28'W], M. Matthews, 13–25.viii, 19–25.ix , 3–12.x.1986 ( NMWC, NMSA) ; YEMEN: 10 _11^, Socotra Archipelago , Socotra, Diksam Mtn. [12 ° 31.40'N: 53 ° 57.20'E], H. Pohl , 26–27.x.2000 ( HLMD, NMSA) .

Synonymy of B. nodosus: Using the excellent description and illustrations published by Rust & Byers (1976), it is my opinion that B. nodosus is a synonym of B. chevalieri . Rust & Byers (1976) recognised that ‘the male genitalia of nodosus closely resemble those of B. chevalieri … especially in the epiandrial lobes’, but noted that ‘ B. nodosus differs … in having a short, spined, dorsal protuberance at the apex of each epiandrial lobe’ and that ‘the posterior margins of the basistyles are abruptly truncate, not rounded’.. While the dorsal protuberance at the apex of the epandrial lobe is slightly different in development when compared with West African males of B. chevalieri ( Fig. 6 ), the condition illustrated by Rust & Byers (1976) appears identical to that seen in males from Socotra ( Fig. 4 ). As far as the posterior margin of the basistyles is concerned, Rust & Byers’s Fig. 9 does demonstrate a somewhat more truncated condition when compared with both the West African and Socotran material. However, the Rust & Byers illustrations appear to have been made from a dry specimen (i.e. not macerated in warm KOH) as details of segment 10 and associated structures (e.g. proctiger) are not illustrated – presumably because they were contracted between surrounding sclerites (as is commonly the condition in dry mounted material). This being true, I suggest that the more rounded appearance of the basistyles of West African and Socotran males may well be largely due to the softening of these structures during maceration. Even if this is not the case, a degree of variation can be anticipated in populations so widely separated (as is evident between the West African and Socotran populations), and so I am confident that future collecting will demonstrate clinal variation and support the contention that nodosus and chevalieri are indeed conspecific.

Validity of previous African records: Male specimens pose no identification problems as the genitalia are quite characteristic. Byers (1971) was satisfied that all previously recorded females were conspecific, with possible exceptions being the female labelled ‘Africa’ reported by Esben-Petersen, because the venation demonstrated slight differences when compared with other specimens, and the female from Djibouti, because the locality was so far removed from Chad and the West African localities. With the discovery of the Socotra specimens, I am prepared to accept that the Djibouti record is probably correct and, with the now documented variation in venation, I am satisfied that the wings illustrated by Esben-Petersen (1921) do represent B. chevalieri . While additional material would assist in supporting these views, I accept all the material listed in this paper as being correctly identified.

Notes on previous illustrations: Three authors published illustrations of one or both wings, and Byers (1971) illustrated the male genitalia of B. chevalieri . Most of these illustrations are defective in one or other respect. Navás’s (1908) drawing of a single fore-wing shows most of the essential characters of the species, but is defective in that no Scv or Cuv is shown, and the wing tip has a somewhat pointed appearance. All the wings I have studied have Scv and Cuv present and have rather more rounded apices. Esben-Petersen’s (1921) photograph of both fore- and hind-wing is good and shows that the drawing he published in 1913, of the same specimen, was defective in that the Scv was omitted. Byers’s (1971) drawings of the male terminalia are generally accurate, although no details of segment 10 and its associated structures are given, apart from a gonostylus which has its base hidden by S9. These drawings were probably made without first removing and macerating the terminalia.

Closely related species: Byers (1971) briefly noted that B. chevalieri is similar to both B. aequalis Navás, 1914 and B. pobeguini ( Navás, 1908) . The male genitalia of both these species are, however, quite different from B. chevalieri (see Byers (1971) and Londt (1972)). Rust & Byers (1976) draw attention to similarities between their B. nodosus and B. latipennis Gerstaecker, 1885 and B. henryi Kimmins, 1928 . B. latipennis is clearly different in a number of respects, but B. henryi is very similar and is probably the closest relative of B. chevalieri .. All these species can, however, be readily distinguished on male genital characteristics.

Distribution: The distribution of B. chevalieri is depicted in Fig. 10 . The eleven known localities all fall within about 14 degrees of latitude (from ca. 11– 25 ° N), a narrow zone when compared to the very extensive range in longitude of some 85 degrees (from ca. 13 ° W – 72 ° E). With a distance of about 9000 km between its Senegal and Indian populations, this species has what must be the most extensive distribution of any mecopteran. It is particularly interesting that all the known localities fall within two or three major natural vegetation categories (depending on the maps you use) dominated by grass; namely ‘Savanna’ (Grass and Scrub), ‘Steppe’ (Short Grass) and ‘Desert Vegetation’ (Xerophytic Shrub, Grass and Cactus) (terminology that of The Times Atlas of the World – Comprehensive Edition 1988). It can be anticipated that any species adapted for existence in grassland is likely to be very widely distributed. Maps suggest that the Indian population of B. chevalieri coincides with the eastern limits of the grassland biome at that latitude, and so it is not anticipated that future collecting will extend the distribution very much further.

Phenology: Except for the Indian material, collected in January, all other samples were collected during August, September and October – a period corresponding to the northern hemisphere’s late summer and autumn, when the development of grasses is probably at its peak.

Figures

Figs 1–3. Bittacus chevalieri (Navás, 1908). 1. Head, anterior aspect. 2. Fore-wing. 3. Hind-wing. Scale lines = 1 mm, all material from Socotra (NMSA). Abbreviations: A 1 – First Anal vein, Cuv – Cubital crossvein, FM – First fork of the Media, FRs – First fork of the Radial Sector, Pcv – Pterostigmal crossveins, Scv – Subcostal crossvein.

Figs 4–9. Bittacus chevalieri (Navás, 1908), genitalia. 4–5. Male from Socotra. 6–7. Male from Mali (Mourdiah). 8–9. Female from Socotra. Scale lines = 1 mm, all material in NMSA.

Fig. 10. Bittacus chevalieri (Navás, 1908), distribution.