Introduction

The Mydidae is one of the smaller fly families, distributed nearly worldwide, with currently almost 500 described species in all (Courtney et al. 2017; Dikow 2017). Kirk-Spriggs and Stuckenberg (2009: Table 6.2) predicted that alone in the Afrotropical fauna about 200 species may be discovered additionally. Mydid flies inhabit generally warm and arid or semi-arid climates. The biology and feeding habits of adult mydid flies are not well known. The adults visit flowers and are most probably nectarophagous (Larson et al. 2001; Orfinger et al. 2018) or/and pollen-feeders (Dikow 2017).

Dikow et al. (2017: Fig. 7) dated the separation of the Mydidae from their sister group Apioceridae at 135 million years ago. These mostly rather large flies are rare in the fossil record, with only two described taxa, viz. the Aptian Cretomydas santanensis Willkommen, 2007 (Crato, Brazil), and the late Eocene Mydas miocenicus Cockerell, 1913 (Florissant, Colorado, USA) (Cockerell 1913; Willkommen and Grimaldi 2007; Pouillon and Nel 2020). The latter would need a revision but the holotype is poorly preserved (see remarks on this fossil below).

Here we describe a third fossil mydid fly, originating from the Paleocene maar lake of Menat (France).

Material and methods

The middle Paleocene Menat site, near the south of the village of Menat (46°06.282’N, 002°54.370’E, Menat Basin, Puy-de-Dôme, France), is a former maar lake (ca. 1 km in diameter), resulting from a phreatomagmatic eruption, filled with organic-rich sediments. This palaeolake contains numerous remains of a diverse aquatic and terrestrial flora and fauna (Piton 1940; Nel 1989, 2007; Nel and Roy 1996; Nel and Auvray 2006; Nel et al. 2010; Garrouste et al. 2017; Schubnel and Nel 2019; Schubnel et al. 2020; Nel and Jouault 2022; Nel and Kundura 2023). The composition of faunal and floral remains suggests that this lake was surrounded by a forest and that the palaeoenvironment was warm and humid (Wedmann et al. 2018). Following the pollen, mammalian stratigraphy, and radiometric K/Ar analyses, its age was estimated as 56 Ma (Kedves and Russell 1982); vs. an age within 60–61 Ma, based on macroflora (Wappler et al. 2009) and dipteran insects (Nel 2007). Flies were quite diverse at the time but are, probably due to taphonomic reasons, uncommon in the Menat deposit. They represent only about 2% of the very rich and diverse taphocoenosis, which is dominated by Coleoptera and Blattodea (Piton 1940; Wedmann et al. 2018).

The fossil specimen was examined with a Leica MZ12.5 stereomicroscope, the drawings were done with an attached camera lucida. The photographs were taken with a Leica MZ12.5 stereomicroscope and an attached Nikon D300 digital camera.

We follow the body and wing venation nomenclature of Dikow (2017).

Systematic palaeontology

Order Diptera Linnaeus, 1758

Superfamily Asiloidea Latreille, 1802

Family Mydidae Latreille, 1809

Genus Paleomydas gen. nov.

Type species: Paleomydas menatensis sp. nov.

Etymology: Named after the Paleocene period and the genus name Mydas. Gender neutrum.

Diagnosis: The main apomorphic character is the very broad hind femur, only about 2.7 times as long as broad; other crucial characters are: venation of genus Mydas type, viz. two submarginal cells; a M3 + 4 veinlet; vein M1 + 2 is separated from the radial veins; R4 joining common R vein before wing margin; R5 ending on anterior wing margin at apex of common R.

Paleomydas menatensis sp. nov.

(Figs. 1, 2)

Fig. 1
figure 1

Photographs of holotype of Paleomydas menatensis gen. et sp. nov. a Overview of plate A. b Overview of plate B. Both scales = 5mm

Full size image
Fig. 2
figure 2

Photographs of wing of holotype of Paleomydas menatensis gen. et sp. nov. a Detail of wing from plate A. b Detail of wing from plate B. c Reconstruction of wing venation from both plates, with labeling of important veins and cells. Scale same for all images = 2mm

Full size image

Etymology: Named after the type locality Menat. The specific epithet is to be treated as an adjective.

Material: Holotype MNT-14-7535 a-b (part and counterpart of an adult fly with body partly preserved, showing incomplete head, thorax and abdomen, with mid and hind legs), stored at the Musée de Paléontologie of Menat¸ Menat, Puy-de-Dôme, France. Figured also in Wedmann et al. (2018: Fig. 13B).

Type locality: Menat, Puy-de-Dôme, France. Collected at outcrop «Stream site», 46°06.282’N, 002°54.370’E, as illustrated in Wedmann et al. (2018, Fig. 1).

Type horizon: Middle Paleocene, Menat Basin.

Diagnosis: As for the genus.

Description: Fossil split into two plates; plate A showing mostly ventral structures (Fig. 1a), plate B showing mostly dorsal structures (Fig. 1b). Body preserved fragmentarily, only right wing preserved. Head very fragmentary; thorax minimum 7.5 mm long (incompletely preserved, plate B), ca. 8.0 mm wide (as preserved on plate B); longitudinally directed setae/bristles present on lateral margin of scutum (plate A) but no visible bristles on posterior margin of scutellum; abdomen as preserved on plate A ca. 11 mm long, abdomen anteriorly ca. 7 mm wide as preserved on plate A; mid femur ca. 4.4 mm long, ca. 1.0 mm wide, mid tibia ca. 4.2 mm long, ca. 0.7 mm wide; both femur and tibia of middle leg covered with setae, perhaps row of longer setae present on tibia; hind femur very broad, ca. 6.5 mm long, ca. 2.4 mm wide, about 2.7 times as long as broad, with long bristles; hind tibia ca. 4.5 mm long, 0.9 mm wide, covered with rather long bristles, perhaps with a longitudinal carina; wing 15.8 mm long, 5.6 mm wide (plate B), hyaline, without visible corrugation; d cell 6.9 mm long, max. 1.2 mm wide (plate B); cell m3 5.0 mm long, max. 1.1 mm wide (plate B); cell r1 between R1 and R2 9.2 mm long, 0.7 mm wide (plate A); R4 and R5 mostly parallel, curved; R4 joining common R vein before wing margin; stump vein (R3) emerging from R4 near its base; R5 ending on anterior wing margin at apex of common R; M1 and M2 fused, M1 + 2 separated from the radial veins; M3 nearly straight; M4 curved; distal M3 + 4 as a short veinlet connecting to posterior wing margin (C) near apex of cell m3, not aligned with M1 + 2; r-m near apex of d cell, with apical part of vein M1 + 2 making a curve before r-m; cell cua closed near wing margin, petiolate; vein CuP weakly undulate; no anal vein visible; a rather large alula seems to be present, ca. 2.0 mm long, ca. 1.0 mm wide.

Discussion of systematic affinities of Paleomydas menatensis gen. et sp. nov.

Four families of flies have a wing venation similar to that of the new fossil, with veins appearing ‘tangled’, and branches of the medial vein curved forward and convergent with radial branches before apex, viz. the Nemestrinidae, Mydidae, Apioceridae, and the Mesozoic family Protapioceridae. Affinities with the Nemestrinidae are excluded because these flies have a composite diagonal vein straight from the discal cell to the outer wing margin, which is not the case in the new fossil. The Apioceridae are also excluded because the discal cell and the cell m3 are narrow in Cretomydas. The Protapioceridae have also a discal cell broader and shorter than in Cretomydas (Ren 1998; Zhang et al. 2007; Zhang 2015). The new fossil differs from the Apioceridae and Protapioceridae in the presence of a vein M1 + 2, a character present in the majority of Mydidae. The Protapioceridae and the Apioceridae have two separated branches M1 and M2, currently considered as a plesiomorphy in Apioceridae and Mydidae (Artigas and Papavero 1990; Yeates and Irwin 1996). The discal cell d and the cell m3 are narrower in the new fossil than in the Apioceridae.

Yeates and Irwin (1996) made the only phylogenetic analysis of Apioceridae and Mydidae on the basis of morphological characters and proposed a series of synapomorphies of the Mydidae, the majority of which are not discernable in the new fossil. Nevertheless, the Apioceridae have long, strong bristles on the lateral margins of the scutum and on the posterior margin of the scutellum, unlike in the new fossil; the absence of these setae is considered as a synapomorphy of the Mydidae. Thus we consider the new fossil as belonging to the Mydidae.

Some extant Mydidae have their two veins M1 and M2 separated, excluding affinities of the new fossil with these taxa, viz. the three genera Diochlistus Gerstaecker, 1868 (Australia), Mitrodetus Gerstaecker, 1868 (Argentina and Chile) which Papavero and Wilcox (1974) and Artigas and Papavero (1990) considered as a subfamily Diochlistinae Bequaert, 1961, plus Rhaphiomidas Osten Sacken, 1877, a taxon originally in Apioceridae, but currently in the mydid subfamily Rhaphiomidinae (Yeates and Irwin 1996; Ovtshinnikova 2003).

The vein R4 joining common R vein before the wing margin is a synapomorphy allowing the attribution of the new fossil to the Mydidae minus Rhaphiomidas Osten Sacken, 1877 (unique genus in subfamily Rhaphiomidinae Williston, 1893). This genus also has CuA and CuP not fused well basad the wing margin, unlike in the new fossil.

The r-m crossvein of the new fossil is close to the apex of the discal cell and at an acute angle to the adjoining proximal region of M, as an apomorphy of the Mydidae except Megascelinae. Vein M1 + 2 is separated from the radial veins in the new fossil, a plesiomorphy, unlike in the Megascelinae, Nemomydas Curran, 1934 (in Leptomydinae), and Syllegomydas Becker, 1906 (Syllegomydinae), in which M curves forward to fuse with the R veins (Yeates and Irwin 1996, Carles-Tolrá 2014). Thus we consider that the new fossil belongs to the ‘remaining Mydidae’ (sensu Yeates and Irwin 1996: fig. 64), excluding the Rhaphiomidinae and Megascelinae (= currently the three genera Megascelus Philippi, 1865, Neorhaphiomidas Norris, 1936, and Tongamya Stuckenberg, 1966) (see also Norris 1936; Stuckenberg 1966).

In the Mydinae, Syllegomydinae, Leptomydinae, Cacatuopyginae, and the ectyphine genera Ectyphus Gerstaecker, 1868, Opomydas Curran, 1934, and Parectyphus Hesse, 1972 (Curran 1934; Hesse 1972; Wilcox and Papavero 1971; Lyons and Dikow 2010), R5 curves forward to join R2 + 3+4 at the point where this vein ends in the costa or ends on R2 + 3+4 (apomorphic state), as in the new fossil. It is not the case in the Rhopaliinae (Perissocerus Gerstaecker, 1868, Pseudorhopalia Wilcox and Papavero, 1971, Rhopalia Macquart, 1838, and Rhopaliana Séguy, 1934; see Calhau et al, 2014) and the genera Anomalomydas Papavero and Wilcox, 1974 (unique genus of Anomalomydinae Papavero and Wilcox, 1974, also charactererized by the cell cua widely open, unlike in the new fossil) (Papavero and Wilcox 1974), the Apiophorinae Papavero and Wilcox, 1974 (Apiophora Philippi, 1865, Eumydas Wilcox and Papavero, 1971, Midacritus Séguy, 1938, Miltinus Gerstaecker, 1868, and Paramydas Carrera and d’Andretta, 1948) (see Séguy 1938; Norris 1938; Carrera and d’Andretta 1948; Wilcox and Papavero 1971; Papavero and Wilcox 1974), and the genus Heteromydas Hardy, 1944 (Ectyphinae Wilcox and Papavero, 1971) (Hardy 1944). Thus we exclude affinities with these taxa.

The Syllegomydinae (Afroleptomydas Bequaert, 1961, Afromydas Bequaert, 1961, Agaperemius Hesse, 1969, Arenomydas Hesse, 1969, Cephalocera Latreille, 1829, Cephalocerodes Hesse, 1969, Eremohaplomydas Bequaert, 1959, Halterorchis Bezzi, 1924, Haplomydas Bezzi, 1924, Heleomydas Séguy, 1929, Heteroleptomydas Bequaert, 1963, Hispanomydas Arias, 1914, Lachnocorynus Hesse, 1969, Mahafalymydas Kondratieff, Carr and Irwin, 2005, Mimadelphus Hesse, 1972, Mydaselpis Bezzi, 1924, Namadytes Hesse, 1969, Namamydas Hesse, 1969, Namibimydas Hesse, 1972, Neolaparopsis Hesse, 1969, Nomoneura Bezzi, 1924, Nomoneuroides Hesse, 1969, Nothomydas Hesse, 1969, Notosyllegomydas Hesse, 1969, Oreomydas Hesse, 1969, Syllegomydas Becker, 1906, Vespiodes Hesse, 1969) have no short vein M3 + 4, excluding affinities of the new fossil with this subfamily (Hesse 1969, 1972). This is also the case for Lampromydas Séguy, 1928 and Dolichogaster Macquart, 1848 (Séguy 1928: 133).

The Leptomydinae genera Eremomidas Semenov, 1896, Hessemydas Kondratieff, Carr and Irwin, 2005, and Leptomydas Gerstaecker, 1868 also have no short vein M3 + 4 (Dikow 2017).

The venation of the Cacatuopyginae (Charimydas Bowden, 1984, Cacatuopyga Papavero and Wilcox, 1974) strongly resembles that of the new fossil, but the hind tibiae of the new fossil are probably carinate, while they are cylindrical in Cacatuopyginae (Papavero and Wilcox, 1974). Also Charimydas and Cacatuopyga differ from the new fossil in having slender femora (Papavero and Wilcox 1974; Bowden 1984) while they are very broad in the latter.

The venations of the leptomydine genera Pseudonomoneura Bequaert, 1961, and Nemomydas Curran, 1934 strongly resemble that of the new fossil. But the latter has very broad hind femora, unlike in these genera (Curran 1934; Hardy 1950; Bequaert 1961).

The ectyphine genus Parectyphus differs from the new fossil in the presence of three submarginal cells, instead of two as in the majority of Mydidae (Hesse 1972: 166). Parectyphus has markedly thickened hind femora as in Ectyphus, but less than in the new fossil (Lyons and Dikow 2010). Opomydas has thin hind femora (Curran 1934: 165; Wilcox and Papavero 1971).

Within the Mydinae, affinities with Protomydas Wilcox, Papavero and Pimentel, 1989 are excluded because the latter has the ventral keel of the hind tibia underdeveloped (Wilcox et al. 1989). Unfortunately, the presence (in Stratiomydina) vs. absence (in Mydina) of a pollinose pattern of stripes or spots on the mesonotum cannot be decided.

Plyomydas Wilcox and Papavero, 1971 has narrower hind femora than the new fossil (Wilcox and Papavero 1971; Castillo and Dikow 2017).

Gauromydas Wilcox, Papavero and Pimentel, 1989, Belemiana Ponting, 2021 (= replacement name for Utinga Wilcox, Papavero and Pimentel, 1989 in Ponting (2021)), Chrysomydas Wilcox, Papavero and Pimentel, 1989, Baliomydas Wilcox, Papavero and Pimentel, 1989, and Messiasia d’Andretta, 1951 have hind femora distinctly thinner than in the new fossil (d’Andretta 1951; Wilcox et al. 1989). The species of the genus Mydas Fabricius, 1794 have their hind femora more than 5 times as long as broad (Wilcox et al. 1989). Some species of Stratiomydas Wilcox, Papavero and Pimentel, 1989 have rather swollen hind femora but less than in the new fossil (five times as long as wide or more vs. 2.7 times in the latter) (Wilcox et al. 1989).

Mapinguari Papavero and Wilcox, 1974 has ‘hind femora exceptionally strongly swollen, 3.8 times as long as broad’, similarly to the situation in the new fossil (Papavero and Wilcox 1974: 53; Calhau et al. 2016). The species of Ceriomydas Williston, 1898 also have swollen femora, from 3.5 to 5.5 times as long as broad, with tuberculate spines on venter, as in the new fossil (Papavero and Wilcox 1974: 39). The species of Phyllomydas Bigot, 1880 have the hind femora between 4.1 and 6.0 times as long as broad, which would be compatible with the situation in the new fossil (Wilcox 1978). The differences between these genera concern characters that are not available in the new fossil (e.g. antennae, genitalia). Also, it seems that the shape of the hind femur is rather variable within a particular mydine genus. Thus, it seems not possible to define the exact affinities of the new fossil to a genus within this subfamily. Altogether, we cannot assign this new fossil to a particular mydine subfamily with confidence. Therefore, we attribute it to the new genus Paleomydas gen. nov. in the hope that it can be placed properly in the future.

Remarks on Mydas miocenicus Cockerell, 1913 (Fig. 3)

Fig. 3
figure 3

Photograph of holotype of Mydas miocenicus Cockerell, 1913, stored in the Collection for Invertebrate Paleontology at Museum of Natural History, University of Colorado, Boulder, Colorado (USA). a Overview of fossil with its cracked resin covering. b Enlarged fossil wing, rotated counterclockwise 20 ° in comparison to Fig. 3a, longitudinal axis of wing parallel to top photo margin; contrast enhanced. Only parts of the wing venation can be discerned. White arrows indicate preserved wing base. Photographs courtesy Talia Karim, Boulder. Both scales = 1mm

Full size image

The holotype of M. miocenicus which is stored at the Collection for Invertebrate Paleontology at Museum of Natural History, University of Colorado, Boulder, Colorado (USA), is in difficult conservatorial condition (see Fig. 3a, the fossil is illustrated for the first time). Formerly many of the historical Florissant specimens were coated with some kind of resin, which has subsequently yellowed and cracked (personal information Talia Karim, Boulder, to SW, 2016). Due to this, seeing the fossils and taking photographs of them is challenging and difficult, and even after editing of the photographs not all mentioned details from the description by Cockerell (1913) can be seen (Fig. 3b). A revision of the fossil species might be possible after careful re-investigation of the original, which was not possible in the scope of this investigation. But it is not predictable how much information can be deciphered from the original fossil, and since it is only part of a wing, no well-founded attribution to a genus will be possible. The long and curved veins M3 and M4 together with the short vein M3 + 4 between M4 and the posterior wing margin, plus the elongate cell m3 are discernable in the photograph, but crucial characters like the radial veins and M1 + 2 are not visible.

Conclusion

Paleomydas menatensis gen. et sp. nov., the third described fossil representative of the Mydidae, shows more similarities with the New World Mydinae than with the Mydidae of the rest of the World. The most similar taxa are the Neotropical genera Mapinguari and Ceriomydas, and the Nearctic genera Phyllomydas and Mydas. The presence of a Mydidae with putative New World affinities in the Paleocene of France is not so surprising as some other insects with New World affinities are known from the European Paleogene. Examples for this are the dysagrionid damselfly genus Petrolestes (Garrouste and Nel 2015), the stink bug genus Eospinosus (Wedmann et al. 2021), and giant ants of the genus Titanomyrma (Archibald et al. 2023), which support the existence of ‘routes’ of migration between South America, North America and Europe during this whole period (e. g. Garrouste and Nel 2019; Archibald et al. 2023). Fossil mantidflies (Neuroptera: Mantispidae: Symphrasinae) belonging to a group which today only occurs in the Neotropical and southernmost Nearctic regions, were shown to be present in Europe in the Paleogene (Wedmann and Makarkin 2007).

Mydid larvae are predators on insect (in particular beetle) larvae and pupae living in the soil, stump, logs, termite, ant or mammal nests (Wilcox 1983; Dikow 2017). Ants and termites are present and large beetles are quite diverse in the insect assemblage of the Menat paleolake.

The discovery of a mydid fly in the insect assemblage of Menat is rather surprising for two reasons: first the paleoclimate of Menat is currently considered to have been warm but humid, corresponding to an evergreen forest surrounding a rather small lake, while the extant mydids are generally living in rather dry habitats; second, flies are not frequent at all in this assemblage, with less than 2% collected specimens (Wedmann et al. 2018). For instance, the Bibionidae are quite rare with three known specimens while this family is often very frequent in the European Paleogene. Thus, the discovery of a Mydidae, a relatively infrequent fly family nowadays, would appear curious. Possibly this family was more frequent during the Cretaceous-Paleocene than today.