The new genus and species is known from a single adult preserved in a clear piece of Burmese amber, which is derived from amber deposits named Noije Bum, near Tanai in the Hukawng Valley of northern Myanmar. A general introduction of the amber deposit and its geological settings can be found in Grimaldi et al. (2002) and Ross et al. (2010). Recent U-Pb zircon dating constrained the Burmese amber to a maximum age of 98.79 ± 0.62 Ma (Earliest Cenomanian; Shi et al., 2012). The type specimen is deposited in the Nanjing Institute of Geology and Palaeontology, Chinese Academy of Sciences, Nanjing, China. Observations and photographs were made using a Zeiss Discovery V20 stereo microscope and a Zeiss Axio Imager 2 light microscope with a digital camera attached respectively. Photomicrographs with green background are taken using green fluorescence as light source attached to a Zeiss Axio Imager 2 light microscope. For the descriptions of thoracic structures, we follow the terminology of Zaragoza Caballero (1991), although certain structures are doubted by Batelka et al. (2011).

Order Coleoptera Linnaeus, 1758

Family Ripiphoridae Gemminger and Harold, 1870 [1853]

Subfamily Ripidiinae Gerstaecker, 1855

Tribe Ripidiini Gerstaecker, 1855

Genus Protoripidius gen. nov.

Type species. Protoripidius burmiticus gen. et sp. nov.

Diagnosis. The genus (based on male) can be separated from all other ripiphorids by a combination of the following characters: antennae filiform, with ten antennomeres; scape and pedicel short, third antennomere elongate; eyes large, not forming unbroken ring around head; mouthparts reduced; mandibles not developed; maxillary palpi three-segmented, gradually shortened and narrowed toward apex; labrum elongate; pronotum subtriangular, lacking grooves or impressions; metanepisternum broad, lacking anterior lobe; elytra short, each elytron distant from one another for the entire length; abdomen with seven visible sternites; all tibial spurs absent; tarsal formula 5-5-4; and first metatarsomere shorter than three remaining combined.

Etymology. The new genus-group name is a combination of the Greek πρωτο- (prōto-), meaning ‘first’, and Ripidius, the type genus of the extant tribe Ripidiini. The name is masculine in gender.

Remarks. Protoripidius clearly belongs to the superfamily Tenebrionoidea due to its characteristic 5–5–4 tarsal formula. The new genus is placed in Ripiphoridae based on its general body habitus (except the filiform antennae) similar to many brachypterus members of the family, such as Ripidiinae and Ripiphorinae. We include Protoripidius in the subfamily Ripidiinae based on the following combination of characters: •

abdominal sternite II visible externally, similar to sternite III;

Currently, the Ripidiinae comprise two extant tribes: •

Eorhipidiini;

Eorhipidiini can be separated from Ripidiini by the unmodified antennomeres, well-developed mandibles, metanepisternum lacking an anterior lobe and abdomen with six visible sternites (Falin and Engel, 2010). Protoripidius appears to intermingle characters of these two extant tribes. It shares with Eorhipidiini the filiform (unmodified) antennae, metanepisternum lacking an anterior lobe and well-developed labrum (although labrum is absent in some Pterydrias species); with Ripidiini the absence of mandibles, three-segmented maxillary palpi (four-segmented in Eorhipidiini), each elytron distant from one another for the entire length and abdomen with seven visible sternites. It is plausible that the new genus likely represents a stem group of the derived Ripidiini and we tentatively place Protoripidius in Ripidiini based on the putative synapomorphies of the tribe, including the highly reduced mouthparts and the widely separated elytra.

Protoripidius burmiticus sp. nov.

(Fig. 1 and Fig. 2).

Etymology. The specific epithet burmiticus refers to the occurrence of the new species in burmite (Burmese amber).

Material. Holotype, NIGP166155, male; lowermost Cenomanian, Hukawng Valley, northern Myanmar; deposited in the Nanjing Institute of Geology and Palaeontology, Chinese Academy of Sciences, Nanjing, China.

Diagnosis. As for the genus (vide supra).

Description. Male. Body small, 2.85 mm long, slightly narrowed anteriorly and posteriorly; dorsum slightly convex. Vestiture consisting of short hairs, not forming pattern.

Head (Fig. 2A) rounded, width 0.64 mm across eyes, length 0.56 mm from posterior margin of head to apex of labrum, slightly declined, with mouthparts directed ventrally, gradually narrowing posteriorly to form narrow neck and short temples. Occipital region without median endocarina; transverse occipital ridge broadly rounded and above plane of pronotum, abutting anterior edge of pronotum and concealing neck from above. Eyes (Fig. 2A, B) slightly emarginate and hypertrophied, but separated both dorsally and ventrally, coarsely facetted without interfacetal setae. Antennal insertions slightly concealed from above, narrowly separated; subantennal grooves absent. Labrum (Fig. 2C) elongate, anterior margin slight curved. Antennae (Fig. 2B) 10-segmented, filiform; antennomere 1 short, 0.11 mm long; antennomere 2 shorter than antennomere 1, 0.08 mm long, apical margin oblique; antennomere 3 longest, 0.31 mm long; antennomeres 4–9 almost in same width, length of antennomeres 4–10 being 0.13 mm, 0.15 mm, 0.14 mm, 0.15 mm, 0.13 mm, and 0.14 mm; antennomere 10 fusiform, 0.25 mm long. Mandibles and maxillae absent; maxillary palpi (Fig. 2C) 3-segmented, palpomere 1 broad and longest, 0.18 mm long; palpomere 2 slightly narrower and shorter than palpomere 1, 0.11 mm long; palpomere 3 short, 0.09 mm long, pointed at apex.

Pronotum (Fig. 2A) transverse, widest at base and narrowing anteriorly, 1.03 mm wide and 0. 66 mm long. Posterior edge sinuate. Prosternum in front of coxae shorter than mid length of procoxal cavity, apparently with median groove. Prosternal process absent. Procoxae long and projecting. Mesonotal disc (Fig. 2D) rectangular, posterior margin slightly sinuate; metascutellum (Fig. 2D and H) large, subquadrate. Elytra (Fig. 2D, E and H) short, 0.91 mm long and each 0.41 mm wide, exposing much of hind wing, abdomen and pterothorax, widest near middle; elytral apices rounded; disc not costate; epipleura absent. Mesocoxal cavities open laterally. Metaventrite convex; exposed portion of metanepisternum (Fig. 2E) broad, lateral sides nearly parallel, with indistinct punctuation; metepimeron (Fig. 2E) broadly exposed, narrower than metanepisternum, with dorsal margin curved. Hind wing (Fig. 2H) exposed, not folded beneath elytra.

Legs slender; trochanterofemoral joint strongly oblique with femora and well-separated from coxae; protibiae longer than protarsi; lacking tibial spurs; all tarsomeres simple; pro- and mesotarsi 5-segmented; protarsomere 1 elongate, as long as protarsomeres 2–4 combined, protarsomere 5 elongate, as long as protarsomeres 3–4 combined (Fig. 2F); mesotarsi 5-segmented, as long as protarsi; metatarsi 4-segmented, metatarsomere 1 elongate, slightly longer than metatarsomeres 2 and 3 combined, metatarsomere 3 slightly longer than metatarsomere 2, metatarsomere 4 elongate, slightly longer than metatarsomere 1 (Fig. 2G). Pretarsal claws simple, not dentate.

Abdomen with seven free ventrites; postcoxal lines absent. Apical margin of terminal sternite distinctly concave.

The most remarkable character of Protoripidius attributes to the filiform antennae. Among all ripidiine males, this type of antennal form is only found in the putative basal tribe Eorhipidiini (Batelka and Hájek, 2009 and Batelka and Hájek, 2010), although similar antennal type also occurs in Ivierhipidius Barclay, an enigmatic Neotropical genus currently not assigned to any known subfamilies (Barclay, 2015). The antennae of typical ripidiine males are usually uniflabellate (Lawrence et al., 2010). However, in Pterydrias Reitter, the sole genus of Eorhipidiini, the antennae possess 11 antennomeres (Batelka and Hájek, 2009), whereas the antennae are 10-segmented in Protoripidius. Up to now, only four Ripidiinae genera have been known to embody 10-segmented antennae in males (Batelka, 2011), i.e. Paleoripiphorus Perrichot, Nel, and Néraudeau from mid-Cretaceous French amber, Cretaceoripidius Falin and Engel from Burmese amber, Olemehliella Batelka from Eocene Baltic amber (Batelka, 2017), and extant genus Blattivorus Chobaut. The antennal morphology of Protoripidius looks quite peculiar, because the filiform type is plesiomorphic (as in Eorhipidiini), while the reduced antennal segmentation is probably a derived character state.

The eyes of Protoripidius appear to represent a transitional form between Eorhipidiini and Ripidiini: the eyes of the males of Ripidiini are expanded, almost holoptic (forming unbroken ring around head, apomorphy), whereas they are not expanded but hemispherical in Eorhipidiini (plesiomorphy). The mouthparts of Protoripidius appears to be transitional as well: the mandibles of Protoripidius are absent as in the Ripidiini, whereas its labrum is present, unlike most members of Ripidiini. Among Ripidiinae, a labrum is only present in Pterydrias (Eorhipidiini), Riekella Selander, Paleoripiphorus and the new genus Protoripidius. Many other highly derived characters of Protoripidius, such as expanded mesoscutellar shield, widely separated elytra, and 7-segmented abdomen, are supportive of the placement of the genus in the extant Ripidiini, rather than the Eorhipidiini.

Prior to this study, the Burmese amber has yielded three genera of Ripiphoridae: Flabellotoma Batelka, Prokop and Engel as the Mesozoic representative of the primitive subfamily Pelecotominae (Batelka et al., 2016b); and Cretaceoripidius (Cockerell) and Paleoripiphorus, both belonging to the Ripidiinae (Batelka et al., 2016b). Our discovery of a new genus reveals a potentially high palaeodiversity of Ripidiinae in the Cretaceous.

All extant ripiphorids have a life cycle known as endoparasitic on immature stages of other insects at some point in their development (Lawrence et al., 2010). Members of ripidiines parasitize nymphs of cockroaches (Lawrence et al., 2010 and Riek, 1955) and Eorhipidiini might be parasitoids of termites (Batelka and Hájek, 2009). The morphologically specialized Protoripidius from mid-Cretaceous likely had a similar life style to their extant counterparts. Possible hosts of Protoripidius, such as cockroaches or termites, have also been known from the same amber deposit in northern Myanmar (e.g., Engel et al., 2016 and Šmídová and Lei, 2017).

Our discovery of an interesting new ripidiine genus from the Upper Cretaceous amber of Myanmar adds to the palaeodiversity of Ripiphoridae in the Cretaceous ecosystem. Protoripidius bears a number of transitional features that shed new light on the relationships of the two extant tribes and early evolution of Ripidiinae. It also suggests that Rhipidiinae had already diversified and comparatively widespread by at least the mid-Cretaceous, as evidenced by recent fossil discoveries (Batelka et al., 2016b, Falin and Engel, 2010 and Perrichot et al., 2004). Protoripidius bears many important morphological characters which would be helpful for future phylogenetic reconstruction of the relationships of the subfamily Ripidiinae.