New Genera and Species of the Family Throscidae (Coleoptera: Elateroidea) in Mid-Cretaceous Burmese Amber

Simple Summary Throscidae is a relatively small lineage in the beetle superfamily Elateroidea. The Mesozoic fossil records of this family are sparse. Here we describe three new throscid species found in mid-Cretaceous Burmese amber, all represented by well-preserved specimens. These newly discovered species suggest that Throscidae had a high diversity in the Cretaceous. Abstract Captopus depressicepsgen. et sp. nov., Electrothroscus yanpingaegen. et sp. nov. and Pseudopactopus robustus gen. et sp. nov. are reported from the mid-Cretaceous Burmese amber. These new findings greatly extend the Mesozoic diversity of Throscidae, which implies a high degree of morphological disparity for this family in the Cretaceous.


Introduction
The family Throscidae is one of the relatively small lineages of elateroids, with five extant genera and about 150 extant species [1]. Throscids are characterized by their unusual antennal grooves running by the notosternal suture and extending posteriolaterad along the hind margins of hypomera, which could also be a possible apomorphy of this family [2,3]. Historically, Throscidae also included genera such as Drapetes Dejean and Lissomus Dalman (e.g., [4][5][6]), which were later moved into Elateridae, as supported by both morphological and molecular studies [7][8][9]. Throscidae appears to be closely related to the elateroid families Eucnemidae and Cerophytidae, but the relationship among them is far from being settled. Analytical phylogenetic studies have suggested Throscidae to be the sister group of Lissominae + Thylacosterninae [10], Eucnemidae [11], Cerophytidae [9], or Eucnemidae + Cerophytidae [8,12], or an independent lineage [9].
Most fossils of Throscidae were discovered in amber deposits, including Lebanese amber [2] and Burmese amber [3,13] of the Cretaceous, and Oise amber [14] and Baltic amber [3,15] of the Oligocene to Eocene. Despite the great diversity of fossil insects preserved in the mid-Cretaceous Burmese amber [16,17], only two throscid species has been reported in this material to date [3,13]. Here, we report another three members of Throscidae from Burmese amber, which greatly enrich our knowledge on the early diversity of this family.

Materials and Methods
The Burmese amber specimens studied here are derived from amber mines near Noije Bum Village (26 • with a small table saw, ground with emery papers of different grit sizes, and finally polished with polishing powder. Photographs under incident light were taken with a Discovery V20 stereo microscope (Zeiss, Jena, Germany). Widefield fluorescence images were captured with the Zeiss Axio Imager 2 light microscope combined with a fluorescence imaging system. Confocal images were obtained with a Zeiss LSM710 confocal laser scanning microscope. Images under incident light or widefield fluorescence were automatically stacked in Helicon Focus 7.0.2 or Zerene Stacker 1.04. Confocal images were automatically stacked and colour-coded in ZEN 2011 or ZEN 2.3 (Blue Edition), or manually stacked in Adobe Photoshop CC. In the multicoloured images, the colours were coded based on z-depth. Images were further processed in Adobe Photoshop CC to enhance contrast. This published work and the nomenclatural acts have been registered in ZooBank, the official registry of Zoological Nomenclature. The LSID for this publication is urn:lsid:zoobank.org:pub:4797AF94-8912-4DCA-B375-A7D25F16B2A6.
Diagnosis. Frons without carina, but with a deep cavity on both sides. Eyes large. Lateral pronotal ridge incomplete. Prosternum with subparallel prosternal carinae. Metaventrite with well-developed mesotarsal grooves. Abdomen with metatarsal grooves extending beyond the posterior margin of the third ventrite.
Remarks. The new genus Captopus is similar to the extant genus Pactopus Horn and the extinct genus Tyrannothroscus Muona in having both well-developed mesotarsal and metatarsal grooves. Captopus shows some crucial differences from other known extant or extinct species in Pactopus. For example, the metatarsal grooves of Pactopus do not extend beyond the posterior edge of ventrite 3, while in Captopus, the metatarsal grooves extend well beyond ventrite 3 and reach the middle of ventrite 4. More importantly, there is a deep cavity on both sides of frons in Captopus, formed by the expansion of the groove around the ridge bordering antennal insertion. This feature is unique among all known extant and extinct throscids. Captopus differs from Tyrannothroscus additionally in having no supraocular ridges, and lateral pronotal ridge incomplete.
Diagnosis. As for the genus. Description. Body elongate, length 2.5 mm, width 0.9 mm.
Abdomen with five connate ventrites; ventrite 5 twice as long as ventrite 4; metatarsal grooves well-developed, arcuate, reaching the middle of ventrite 4.     Etymology. The generic name is derived from the Greek noun "elektron", amber, and the generic name Throscus Latreille, a junior synonym of Trixagus Kugelann. The name is masculine in gender.
Remarks. In Throscidae, only Pseudothroscus Muona is known to lack median head carina and both tarsal grooves [3]. Pseudothroscus is also characterized by its nonparallel prosternal carinae [13]. However, the prosternal carinae of Electrothroscus are subparallel, which is similar to most Throscidae.
Diagnosis. As for the genus. Description. Body elongate, length 2.3 mm, width 0.8 mm.
Pronotal disc about 1.6 times as wide as long along the middle; sides not sinuate, converging anteriorly; posterior angles strongly acute and produced posteriorly; posterior edge bisinuate. Elytra widest behind the base, about 2.2 times as long as wide combined.
Abdomen with five connate ventrites; ventrite 5 twice as long as ventrite 4; metatarsal grooves well-developed, merely extending to the hind margin of ventrite 3.

Discussion
The three throscid species presented in this paper, together with two previous reported species, show a high degree of morphological variation of Throscidae in Burmese amber. Specialized throscids have also been reported in Lower Cretaceous Lebanese amber [2]. Compared to their extant relatives, these Mesozoic throscids seem to be morphologically more diverse. There are only four known extant genera in Throscidae, while six genera have been established based on Cretaceous fossils. An identification key to

Discussion
The three throscid species presented in this paper, together with two previous reported species, show a high degree of morphological variation of Throscidae in Burmese amber. Specialized throscids have also been reported in Lower Cretaceous Lebanese amber [2]. Compared to their extant relatives, these Mesozoic throscids seem to be morphologically more diverse. There are only four known extant genera in Throscidae, while six genera have been established based on Cretaceous fossils. An identification key to

Discussion
The three throscid species presented in this paper, together with two previous reported species, show a high degree of morphological variation of Throscidae in Burmese amber. Specialized throscids have also been reported in Lower Cretaceous Lebanese amber [2]. Compared to their extant relatives, these Mesozoic throscids seem to be morphologically more diverse. There are only four known extant genera in Throscidae, while six genera have been established based on Cretaceous fossils. An identification key to genera in Throscidae is updated in light of the new specimens discovered from Burmese amber (Appendix A). Many interesting features found in Mesozoic throscids have hitherto been unknown in Recent Throscidae (e.g., deep cavities on head in Captopus gen. nov., small and very narrow eyes in Potergosoma, nonparallel prosternal carinae in Potergosoma and Trixagosoma). The divergence between Throscidae and the group of Eucnemidae + Cerophytidae has been dated to Middle Jurassic, approximately 165 Ma [12]. Therefore it is reasonable to expect that Throscidae might have already been highly diversified in late Mesozoic. We hope further findings on Mesozoic throscid fossils could help us understand the early diversification of this family.
Despite the wide distribution of the family, throscids are poorly studied and their taxonomy is not well resolved. Though the monophyly of Throscidae has been supported by recent molecular phylogenetic analyses [9,23], no molecular-based analysis has focused on the inter-generic relationships within Throscidae. A morphology-based phylogenetic analysis of both extant and extinct throscids was performed by Muona [3]. Although two of three equally shortest trees he obtained from the parsimony analysis showed Aulonothroscus as non-monophyletic, he concluded it was reasonable to opt for the tree supporting a monophyletic Aulonothroscus. However, later Li et al. [13] performed an analysis under implied weights with a slightly modified dataset, in which Aulonothroscus appeared to be paraphyletic in the majority-rule consensus tree.
The generic assignment of some species by Muona [3] was also dubious. For example, a throscid from Eocene Baltic amber, Trixagus parvulus Muona, was assigned to genus Trixagus. However, the most important diagnostic feature for Trixagus, the vestigial tarsal grooves, is not visible on the specimen: the position for possible tarsal groove is concealed by its femur. The club-forming antennomeres in T. parvulus seems to be rather symmetrical and moniliform ( Figure 16 in [3]), whereas in Trixagus they should be asymmetrically expanded and serratiform. Besides, we need to be cautious when evaluating the systematic position of insects in preserved in amber, especially when the specimen is not well-preserved. A failure in identifying a certain structure in an amber specimen does not necessarily mean that it is absent [24]. Various decomposition and distortion processes may lead to the misinterpretation of morphological characters. Further better preserved fossils could be helpful for clarifying the accurate position of these species.

Conclusions
Our discovery of Captopus depressiceps gen. et sp. nov., Electrothroscus yanpingae gen. et sp. nov. and Pseudopactopus robustus gen. et sp. nov. in the mid-Cretaceous Burmese amber greatly extends the Mesozoic diversity of Throscidae. These newly discovered species imply a high degree of morphological disparity for this family in the Cretaceous. Further phylogenetic analysis with incorporation of both molecular data and fossil taxa would be helpful for elucidating the early evolutionary history of Throscidae.