Babinskaiidae is an extinct lacewing family of the superfamily Myrmeleontoidea. Hitherto, nine species of seven genera are described from the Lower and mid-Cretaceous. Here a new species of Babinskaiidae is described from Cretaceous Burmese amber, namely Parababinskaia makarkini sp. n. The new species possesses an A2 vein in the hind wing, suggesting that the loss of this vein might not be an autapomorphy of Babinskaiidae. The female of Electrobabinskaia burmana Lu, Zhang & Liu, 2017 is also described for the first time based on two specimens with their abdomens perfectly preserved, exhibiting a specialised sternum VI with paired elongate projections. A brief discussion of female genital characters is provided, which may increase our understanding of the morphology and phylogenetic position of Babinskaiidae.

Mesozoic, Myrmeleontoidea , Neuropterida , phylogeny, taxonomy

The extinct lacewing family Babinskaiidae, belongs to the superfamily Myrmeleontoidea, and is recently considered to form an epifamily Nymphidoidae together with Nymphidae (Makarkin et al. 2017). Adults of Babinskaiidae can be characterised by long filiform antennae, narrowly elongated wings, with features such as trichosors, and presectorial cross veins present in both wings, and absence of forewing oblique vein (i.e., the base of MP2).

Hitherto, Babinskaiidae were only recorded in the Lower Cretaceous of Brazil (Crato Formation) and Russia (Zaza Formation), and the mid-Cretaceous of Myanmar (Martins-Neto and Vulcano 1989a, b; Ponomarenko 1992; Martins-Neto 1997; Lu et al. 2017; Makarkin et al. 2017). Currently, the family contains nine species assigned in seven genera, i.e., Baisonelia Ponomarenko, 1992 from the Lower Cretaceous of Russia; Babinskaia Martins-Neto & Vulcano, 1989, Neliana Martins-Neto, 1992, Parababinskaia Makarkin, Heads & Wedmann, 2017 from the Lower Cretaceous of Brazil; and Burmobabinskaia Lu, Zhang & Liu, 2017, Electrobabinskaia Lu, Zhang & Liu, 2017, and Pseudobabinskaia Makarkin, Heads & Wedmann, 2017 from the mid-Cretaceous of Myanmar. However, many of these species are from compression fossils, and some of them are known only from wing fragments (e.g., Baisonelia vitimica Ponomarenko, 1992 and Neliana impolluta Martins-Neto, 1997, each with only a hind wing preserved). Recent discovery of Babinskaiidae in Burmese amber provides important evidence to understand the morphology, taxonomy, and phylogenetic status of this family owing to the well-preserved specimens (Lu et al. 2017). Nevertheless, known Burmese amber specimens of Babinskaiidae are still scarce.

In this paper, with examination of more specimens of Babinskaiidae from the Burmese amber, a new species of Parababinskaia is reported based on two specimens with both the male and the female described, and the female of Electrobabinskaia burmana Lu, Zhang & Liu, 2017 is also described for the first time. A comparative study on the female genital morphology of Babinskaiidae is presented.

The amber samples described are from the Hukwang Valley in Tanai Township, Myikyina District of Kachin State, Myanmar (Kania et al. 2015). The age of this deposit has been investigated and dated to be 98.8±0.6 million years by U-Pb dating of zircons from the volcanoclastic matrix of the amber (Shi et al. 2012).

The specimens are deposited in the Nanjing Institute of Geology and Palaeontology, Chinese Academy of Sciences, Nanjing, China, while a paratype of the new species herein described is currently housed in the Entomological Museum, China Agricultural University (CAU), Beijing, and will eventually be deposited in the Collection of Xiao Jia in the Century Amber Museum (CAM), Shenzhen.

Photographs and drawing were taken and made using a Zeiss SteREO Discovery V12 microscope system. The figures were prepared with Adobe Photoshop CS6. Terminology of wing venation generally follows Aspöck et al. (1980) and Martins-Neto (2000). Breitkreuz et al. (2017) presented an alternative interpretation on the homology of wing venation in Neuropterida based on vein tracheation. The corresponding abbreviations of the veins based on the nomenclature in Breitkreuz et al. (2017) are given below in the parentheses for comparison. These two venation terminologies differ from each other mainly in homology interpretation and definition of MA, i.e., whether MA is considered to be fused with RP at the wing base. The venation terminology used for Babinskaiidae in Makarkin et al. (2017) is in generally similar to that of Breitkreuz et al. (2017). Terminology of genitalia follows Aspöck and Aspöck (2008).

Abbreviations used for wing veins are:

A (A) anal vein;

C (C) costa;

Cu (Cu) cubitus;

CuA (CuA) cubitus anterior;

CuP (CuP) cubitus posterior;

M (M) media;

MA (RP1) media anterior;

MP (MA+MP) media posterior;

R (R) radius;

RA (RA) radius anterior;

RP (RP) radius posterior;

ScP (Sc) subcosta posterior;

ps presectorial crossveins (i.e., r-mp crossveins).

The present new findings on the Burmese amber babinskaiids provide important information to further understand the morphology and systematics of Babinskaiidae. Makarkin et al. (2017) outlined four apomorphic wing characters to define Babinskaiidae, including the distal origin of RP+MA, the presence of presectorial cross veins in both wings, the single forewing MP, and the reduction of hind wing A2 and A3. The former three characters may be verified as the synapomorphies of Babinskaiidae although the distal origin of RP+MA and the presence of presectorial cross veins are also present in some lineages of Myrmeleontidae and Ascalaphidae. The long hypostigmal cell is another apomorphic character of Babinskaiidae mentioned in Lu et al. (2017), while it is also present in Nymphidae, Nemopteridae, Palaeoleontidae, and Myrmeleontidae. However, the hind wing A2, possible with A3 merged, is present in P. makarkini sp. n. and E. burmana, indicating that the reduction of A2 and A3 may not be the autapomorphy of Babinskaiidae as proposed by Makarkin et al. (2017).

The specialised sternum of abdominal segment VI in the female of E. burmana is remarkable. In the female of this species there is a pair of long digitiform projections, while such projections are not developed in the conspecific males. In light of the absence of these projections in males, this feature probably functions during courtship or mating although it does not belong to the genital segments. Notably, such modification of abdominal segment VI has never been found in Neuroptera. Previously reported sexually dimorphic features on pregenital segments of abdomen in Neuroptera are only known in males, such as the eversible sacs in some species of Nevrorthidae, Osmylidae and Mantispidae, and the hair pencils in some species of Myrmeleontidae, presumably being involved with chemical communication between sexes (New 1989). However, females of some species of Corydalidae (Megaloptera), e.g., Protohermes differentialis (Yang & Yang, 1986), display unusual features on ventral sclerites of abdomen, such as paired long projections on gonocoxites VIII (see Liu and Yang 2006: figs 16–17). In Neuroptera, females of Osmylidae have complex modifications of gonocoxites VIII and gonapophysis IX for grasping males during copulation (Martins et al. 2016). While the above female traits in Babinskaiidae and other groups of Neuropterida are clearly not homologous we cannot exclude that they may similarly function considering their morphological similarities.

Lu et al. (2017) described similar paired projections on sternum VII in the female of Pseudobabinskaia martinsnetoi (Lu, Zhang & Liu, 2017). The segmentation of abdomen cannot be clearly observed due to preservation condition in the holotype female of P. martinsnetoi. However, based on the position of these projections in respect of whole abdomen, we consider that these projections actually belong to the sternum VI in P. martinsnetoi and are homologous with that in E. burmana. Thus, Electrobabinskaia and Pseudobabinskaia might have close phylogenetic relationship by sharing this feature that is apparently apomorphic. In P. makarkini sp. n. a specialised female sternum VI is not present, suggesting that this female trait is not a diagnostic character of the whole family but only for some genera of Babinskaiidae.

The female genitalia of Babinskaiidae consist of paired gonocoxite VIII, gonapophysis VIII (at least present in E. burmana), paired gonocoxite IX, paired ectoprocts with well-developed callus cerci, and subanale (at least present in E. burmana) (see Fig. 7). The paired digitiform lobes in Pseudobabinskaia, interpreted as gonocoxites IX in Lu et al. (2017), are verified to be gonocoxites VIII. This feature appears to be similar to that in some antlion genera (e.g., Nedroledon Navás, 1914; see Aspöck and Aspöck 2008: fig. 149) although it is likely convergently derived in Babinskaiidae and Myrmeleontidae. In Electrobabinskaia and Parababinskaia, the female gonocoxite VIII are less modified.

The presence of subanale (or cataprocessus in New 1981) in Babinskaiidae is also noteworthy. The subanale is a small singular sclerite usually present beneath anus. In Myrmeleontoidea sensu Engel et al. (2018) it is previously reported only in Nymphidae (New 1981). Whether the presence of subanale is apomorphic or plesiomorphic is unknown as it is also present in Psychopsidae (New 1988; Bakkes et al. 2017), a phylogenetically basal family in Myrmeleontiformia that comprises Myrmeleotoidea and Psychopsoidea (Engel et al. 2018).

The phylogenetic position of Babinskaiidae appears to be perplexing with mixture of character states that are shared with Nymphidae, Psychopsidae or Myrmeleontidae. Although Makarkin et al. (2017) deemed the sister group relationship between Babinskaiidae and Nymphidae, Engel et al. (2018) placed Babinskaiidae in a crown group within Myrmeleontoidea, together with Ascalaphidae and Mymeleontidae. In addition, it should be mentioned that the phylogenetic position of Nymphidae is still controversial. Many phylogenetic studies (Winterton et al. 2010; Wang et al. 2017) suggest the basal most position of Nymphidae in Myrmelontoidea sensu Engel et al. (2018). However, a phylogenetic study based on anchored hybrid enrichment data (Winterton et al. 2018) assigned Nymphidae to be the sister group of Ithonidae that is traditionally considered not to be the member of Myrmeleontiformia. Our new finding provides more knowledge on the morphology of Babinskaiidae. However, phylogenetic analysis combining fossil and extant families of Myrmeleontiformia is required in future studies to further elucidate the phylogenetic position of this enigmatic extinct lacewing family.

We thank Mrs. Xiao Jia for kindly offering materials of Burmese amber lacewings for our study. This research was supported by the National Natural Science Foundation of China (Nos. 31672322, 41572010, 41622201, 41688103) and Beijing Natural Science Foundation (No. 5162016). We thank Dr. Shaun L. Winterton and four anonymous reviewers for their valuable comments which improved the manuscript.