Paleolomatia menatensis sp. n.

Named after the Paleocene and the recent genus Lomatia.

Wing venation characters only. R4 + 5 branched, M2 present, anal cell open, i-r cross-vein absent, vein m-cu long and sigmoidal, R2 + 3 arising very close to base of Rs but not at 90°, R2 + 3 very strongly sinuous at apex, with a distinctly pronounced loop, R4 strongly sinuous at apex with a basal appendix, r-m cross-vein entering discal cell very distally, in the outer 80%, cell r5 open, veins M2 and CuA1 well separated, not convergent, with cell m2 between them longer than broad, R5 and M1 apically separated.

Paleolomatia menatensis sp. n (Fig. 1 and Fig. 2).

Wing 8.8 mm long, 2.5 mm wide, hyaline; no visible elongate spine at basicosta; base of Rs 2.5 mm from wing base; R2 + 3 arising very close to base of Rs but not at 90°; R2 + 3 very strongly sinuous at apex, with a distinct loop, Radial Loop Index RLI = 2.2 (sensu Yeates [25]); 90° bend in apex of R2 + 3; R4 + 5 branched; R4 strongly sinuous at apex but with less pronounced apical curvature than R2 + 3; i-r cross-vein absent, but a basal appendix of R4 present; cells r5 open; r-m cross-vein entering discal cell very distally, in the outer 80%; vein M2 present; vein m-cu long and sigmoidal; cell m2 longer than broad, 1.0 mm long, 0.6 mm wide; M2 meeting wing margin closer CuA1 than M1; anal cell cup open; axillary cell narrow and parallel-sided; alula not preserved.

Head and abdomen missing; thorax 5.0 mm long, 3.0 mm wide; legs elongate but not very well preserved, apparent absence of midtibial spurs; tarsal setae elongate and lanceolate.

Holotype MNHN-LP-R.63892, laboratoire de paléontologie, Muséum national d’Histoire naturelle, Paris.

Menat, Puy-de-Dôme, France.

Paleocene, –56 My, Spongo-diatom sediment of a volcano-sedimentary maar.

Named after the small village Menat.

The lack of information concerning the body structures of this fossil renders its attribution to a precise lineage very delicate. The present attribution is only tentative because we lack nearly all the body structures used by Yeates [28] in his phylogenetic analysis. Nevertheless, this fossil has its R2 + 3, very strongly sinuous at apex, R4 strongly sinuous at apex, which are rather infrequent characters, even among bombyliids.

After Zaitzev [29], Paleolomatia gen. nov. falls in the ‘Bombyliidae’ (sensu Zaitzev), to the exclusion of the Mythicomyiidae, Systropodidae, Phthriidae and Usiidae, because of the vein R4 + 5 branching apically, and anal cell closed. Following the phylogenetic analysis of Bombyliidae (sensu lato) of Yeates [28] and the catalogue of Bombyliidae of Evenhuis [6], only some Lomatiinae Schiner, 1868 have the following combination of characters also present in Paleolomatia: R4 + 5 branched, M2 present, anal cell open, i-r cross-vein absent, vein m-cu long and sigmoidal, R2 + 3 arising very close to base of Rs, R2 + 3 very strongly sinuous at apex, with a distinct loop, R4 strongly sinuous at apex, cell r5 open, R2 + 3 is not arising at 90° [12].

Among the lomatiine genera, Paleolomatia differs from Docidomyia White, 1916 and Aleucosia Edwards, 1934 in the lack of i-r cross-vein. It differs from Comptosia Macquart, 1840, Ylasoia Speiser, 1920, Ulosometa Hull, 1973, Oncodosia Edwards, 1934 and Doddosia Edwards, 1934 in its veins M2 and CuA1 well separated, not convergent, and with the cell between them longer than broad, although this character is rather variable in Comptosia [12], [25], [26] and [27]. Paleolomatia differs from Macrocondyla Rondani, 1863 in the r-m cross-vein entering the discal cell very distally, in the outer 80%, instead of the outer fourth to the outer sixth [12]. The genera Brachydemia Hull, 1973, Bryodemina Hull, 1973, Canariellum Strand, 1928, Lomatia Meigen, 1822, Notolomatia Greathead, 1998, Edmundiella Becker, 1915, Ogcodocera Macquart, 1840, and Anisotamia Macquart, 1840 have R2 + 3 and R4 distinctly less sigmoidal [8] and [12]. The fossil lomatiine genus Alomatia Cockerell, 1914 differs from Paleolomatia in the very basal position of the r-m cross-vein, and the veins R2 + 3 and R4 less sigmoidal [5]. Hull [12] considered the genus Protolomatia Cockerell, 1914 as a Lomatiinae, but Evenhuis [6] and [7] synonymized it with the recent tomomyzine genus Paracosmus Osten Sacken, 1877. Paleolomatia differs from it in the more proximal position of the base of R2 + 3 [4] and [12].

The wing venation of Paleolomatia looks similar to that of Peringueyimyia Bigot, 1886, except for the veins R5 and M1 apically separated, and a more pronounced radial loop in the former [12]. They share the presence of a basal appendix of R4 (but also present at least in Anisotamia, and some species of Aleucosia and Canariellum, etc.) [8] and [25].

Among the other fossil Bombyliidae with very sigmoidal veins R2 + 3 and R4, the lomatiine Macrocondyla miranda (Cockerell, 1909), originally described in the genus Megacosmus Cockerell, 1909 [6], has also very sigmoidal R2 + 3 and R4, similar to those of Paleolomatia [5], but it differs from the latter in the more basal position of r-m cross-vein (at the 73% of the discal cell) and the cell m2 between the veins M2 and CuA1, not really longer than broad (after Cockerell [1]). Ylasoia secunda (Cockerell, 1911), originally considered as a Megacosmus 1909 [6], differs from Paleolomatia in the cell m2, very narrow, as in the recent Ylasoia [3]. The genus Alepidophora Cockerell, 1909, currently considered as a Bombyliidae Eclimini [6], has also very sigmoidal R2 + 3 and R4 [5]. It differs from Paleolomatia in its veins, M2 and CuA1, strongly approximating near posterior wing margin [2], [14] and [15].