Berothidae

Wedmann, Sonja, Makarkin, Vladimir N., Weiterschan, Thomas & Hörnschemeyer, Thomas, 2013, First fossil larvae of Berothidae (Neuroptera) from Baltic amber, with notes on the biology and termitophily of the family, Zootaxa 3716 (2), pp. 236-258 : 251-252

Berothidae in Baltic amber and the taxonomic affinities of the described larvae

Until now, two species of adult Berothidae have been reported from Baltic amber, i.e., Proberotha prisca Krüger, 1923 and Whalfera wiszniewskii Makarkin & Kupryjanowicz, 2010.

Proberotha prisca was described from a single specimen, which was not illustrated and now apparently lost. No other specimens have been reported subsequently. However, the original description shows that this species probably belongs to the Nevrorthidae . In particular, the strongly pectinate CuP (with nine short branches in P. prisca ), and the presence of two gradate series of crossveins in the radial to medial spaces are characteristic of Nevrorthidae , not of Berothidae . The features of several specimens of one undescribed species of Nevrorthidae are in general consistent with those of P. prisca (see e.g., Ross 1998: Fig. 133; Scheven 2004: Fig. on p. 74, left).

Therefore, the only genus of Berothidae recorded with certainty from Baltic amber is Whalfera Engel, 2004. This genus belongs with certainty to the Rhachiberothinae, and it is apparently most closely related to the extant African genus Mucroberotha Tjeder (Makarkin & Kupryjanowicz 2010) .

Four other adult berothid specimens known from Baltic amber are not described. Three of these belong probably to one species or a few closely related species (see Bachofen-Echt 1949: Fig. 122; Weitschat & Wichard 1998: Figs. 55a,b; Scheven 2004: Fig. on p. 7). Their venation most closely resembles that of the extant African berothine genus Lekrugeria Navás (U. Aspöck in Scheven 2004, p. 6), but the scapus appears to be much shorter and the long hairs on the body and antennae characteristic of Lekrugeria are not visible in the photographs. A fourth specimen from the MCZ collection is not illustrated; it is only known that this is a female which “has long hypocaudae and is obviously a berothine” (MacLeod & Adams 1968, p. 258). This specimen is not found in the MCZ (P. Perkins, pers. comm. to VM, 2007). Theoretically, it could be identical with one of the above mentioned species because Lekrugeria possesses long hypocaudae (Aspöck & Aspöck 1985). Therefore, all four known adult specimens from Baltic amber could belong to few closely related species of a berothinae genus similar to Lekrugeria .

Larvae A and B belong with high certainty to the Berothinae based on all characters available (e.g., the configuration of the ecdysial cleavage lines; the structure of the thoracic sclerites; three-segmented antennae terminating in a seta; four-segmented labial palps; an ocular area without distinctly visible stemmata). However, the larvae of the subfamilies Trichomatinae and Nosybinae are as yet unknown, and so some uncertainty remains.

Larvae A and B most probably not are conspecific, but very similar. Apart from the difference in the presence/ absence of the visible caudal sucker (see description), larva B has slightly longer antennae; its lateral sutures are inclined toward the anterior part of the frontal sutures at a clearly obtuse angle (not at a right angle); the legs are more slender. If these differences are not intraspecific, the two larvae might belong to closely related species of one berothine genus.

Larvae C and D possess some unique character conditions among Berothidae (e.g., antennae and labial palps are six- or seven-segmented; ecdysial cleavage lines consist of only frontal and coronal sutures (the lateral suture is absent); pronotal sclerites are large and in contact with each other along the midline). Nevertheless, we place these two larvae in Berothidae . There are six neuropteran families whose larvae have straight jaws (mandibulomaxillary stilets), and all are recorded from Baltic amber: Osmylidae , Sisyridae , Coniopterygidae , Berothidae (including Rhachiberothidae ), Mantispidae and Dilaridae (Aspöck 1992) . Affinities of larvae C and D to the families Sisyridae , Coniopterygidae , Mantispidae and Dilaridae may be rejected by various reasons: sisyrid larvae lack labial palps; coniopterygid larvae are very small and have a strongly different appearance; mantispid larvae at third instar are scarabaeiform. Dilarid larvae lack thoracic dorsal sclerites and visible ecdysial cleavage lines; the head is rounded; the antennae are specialized (e.g., Gurney 1947; Ghilarov 1962; Monserrat 1988). In general, they are dissimilar to our larvae C and D. Osmylid larvae resemble larvae C and D in many aspects, e.g., in possessing large pronotal sclerites that touch each other along the midline; long, slender antennae and labial palps; five to six pairs of stemmata; basally broadened maxillae (see Lestage 1921: Fig. 1 ; Withycombe 1925; Kawashima 1957: Pl. 2, Figs. A–C; New 1991: Figs 34.5A, 34.10E, F). However, the head of osmylid larvae is short, rounded and the mandibulomaxillary stilets are at least twice longer than the head capsule length (the head capsule of larvae C and D is elongate, and the mandibulomaxillary stilets are shorter than the head). Moreover, an osmylid larva known from Baltic amber does not differ from extant osmylid larvae in this respect (see Wichard et al. 2009: Fig. 07.04). Therefore, Berothidae is the most probable family affinity of larvae C and D. They share many features of extant berothid larvae: similar general appearance; very similar structure of head capsule, mandibulomaxillary stilets and legs; the presence of distinct ecdysial cleavage lines; general similarity in the structure of antennae and labial palps (long; slender; without specialized segments); the presence of large pronotal sclerites, and smaller ones on the meso- and metathorax.

Although larvae C and D are very similar to each other in many external characters, the ecdysial cleavage lines on the head have different proportions. Another similar larva deposited in the private collection of Thomas Schäfer (Saland, Switzerland; No. 01.03.13) is known to us only by its photograph (Larva E) ( Fig. 11 ). This is also a complete mature larva (approximately 7.5 mm long), mostly obscured by a milky covering. Larva E is similar to larvae C and D; it shares with the latter the same structure of the head, antennae (probably seven-segmented) and pronotal sclerites. These three larvae may well belong to closely related species of the same genus; perhaps they are even conspecific.

The taxonomic affinity of larvae C and D within Berothidae is unclear. They belong to some basal subfamily, but not to Rhachiberothinae, Berothimerobiinae or Nyrminae, the larvae of which are known. Unfortunately, the larvae of the basal subfamily Cyrenoberothinae are unknown. There are also some Cretaceous taxa (including Mesithoninae and Paraberothinae) with which these larvae may be theoretically associated, but these taxa have not been recorded from Baltic amber.

Figures

FIGURE 1. Berothidae indet., larva A, whole larva as preserved. A, photograph. B, drawing (both dorsal view). Scale bar is 0.5 mm.

FIGURE 11. Berothidae indet., larva E. A, whole larva as preserved. B, anterior part of the larva (dorsal view). Scale bars are 1 mm (A, B).