Webspinners in Early Eocene amber from western India (Insecta, Embiodea)

Abstract The family Scelembiidae (Neoembiodea: Embiomorpha: Archembioidea) is recorded from Asia for the first time, based on two individuals preserved in Early Eocene amber from the Cambay Basin, western India. Kumarembia hurleyi Engel & Grimaldi, gen. n. et sp. n., is described, figured, and distinguished from other archembioid genera. The genus shares male genitalic features with scelembiids, otherwise known from South America and Africa.


Introduction
Embiodea are one of the more infrequently encountered and investigated orders of insects. This is unfortunate given their remarkable morphological specializations, most of which relate to the production of and life within silken galleries. For example, the probasitarsus is greatly swollen and encompasses distinctive silk glands from which the galleries are spun. The wings are unique among the flying insects for their great flexibility, permitting individuals to move in reverse through their silken tunnels, but can be made more rigid by pumping haemolymph into distinctive 'blood sinuses', enabling them to gain temporary rigor and permit controlled flight. Females are apterous, while males can be either fully winged or shed their wings, much like termites. Even more fascinating is that where known, all species are gregarious, living in small colonies, much like their putative relatives among the Zoraptera.
The relationship of Embiodea to other orders has been problematic, much like everything pertaining to the phylogeny of webspinners. Among the numerous competing hypotheses, those with the greatest support are a relationship to the Phasmatodea (e.g., Rähle 1970;Terry and Whiting 2005;Kjer et al. 2006;Ishiwata et al. 2011) or the Zoraptera (e.g., Engel and Grimaldi 2000;Grimaldi and Engel 2005;Yoshizawa 2007Yoshizawa , 2011. Considerable work continues documenting the diversity of the order, with many hundreds of new species awaiting description (Ross 2000), and revising hypotheses of relationship based on this growing knowledge of the range of morphological variation observed across this fascinating group. Unfortunately, the fossil record has to date provided minimal insights toward clarifying systematic issues pertaining to embiodean evolution. This is because only nine definitive webspinner species are known from the fossil record (Table 1) and many of these are relatively modern, thereby relating more to questions of Tertiary biogeography than to higher-level branching patterns, many of which likely date from the Early Cretaceous or even Late Jurassic. Given this sparse record, any fossil webspinners are of considerable significance.
Herein we provide the description of a new genus and species of fossil webspinner based on two exceptionally well preserved individuals (Figs. 1, 2) recently recovered from Early Eocene amber of the Cambay Basin in western India. These are the first fossil webspinners from Asia (Table 1) and also the first records of their family, Scelembiidae, from the Oriental Region.  Engel and Grimaldi 2006). The fossil Clothonopsis miocenica Hong and Wang (1987) from the Miocene of China was originally described as a clothodid but is actually a bibionid fly (Zhang 1993).
Eocene (Ypresian) India Pachylembiidae: Sorellembiinae Sorellembia estherae Engel & Grimaldi, 2006 Cretaceous (Albian) Myanmar Notoligotomidae: Burmitembiinae Burmitembia venosa Cockerell, 1919 Cretaceous (Albian) Myanmar *Incertae Sedis* Sinembiidae Sinembia rossi Huang & Nel, 2009 Jurassic (Bathonian) Inner Mongolia, China Juraembia ningchengensis Huang & Nel, 2009 Jurassic (Bathonian) Inner Mongolia, China * This species has been placed in the genus "Lithembia" by Ross (1984) but as noted by Engel and Grimaldi (2006) and Miller (2009) the generic name is a nomen nudum and so we have reverted to Cockerell's original combination for our table. The species very likely does not belong to Embia and the two syntypes (UCM-4421 and YPM-26169) should be re-examined and critically revised (based on photographs of the specimens they would appear to have the primitive condition of basal vein branching as delimited by Szumik (1996) (Engel pers. obs.). ** E.S. Ross (pers. comm. 2010) presently does not consider these to belong to Embiodea and, indeed, the presence of a distinct ovipositor, fully-winged females, absence of probasitarsal modifications (which is not swollen despite the assertion of the authors), absence of a radial blood sinus (indeed, from the figures provided, the presence of any blood sinuses seems to require confirmation), and cerci with three cercomeres exclude the species from the order. These species certainly require revision, as do all compressions presently assigned to Embiodea.
Female: Unknown. Etymology. The new generic name is a combination of Kumar (honoring Dr. Kumar Krishna, faithful colleague and dear friend, as well as the world's leading authority on the systematics of Isoptera), and Embia, type genus of and frequent stem for embiodeans. The name is feminine.

Diagnosis.
As for the genus (vide supra). Description. Male: Total length (excluding wings and antennae, as preserved) 5.3 mm; forewing length (estimated) 5.1 mm, width 1.1 mm; integument generally light brown except darker on head and antenna, finely imbricate and impunctate where evident (based on paratype, integument of holotype slightly wrinkled owing to apparent desiccation and shrinkage of individual). Head length (to apex of labrum) 1.1 mm, width (just posterior to compound eyes) 0.64 mm, head posterior to compound eyes longer than compound eye diameter, posterior border gently rounded, covered with numerous, short, prominent setae, longer ventrally (Fig. 3). Pronotum length 0.56 mm, width (medial) 0.40 mm, apparently with weak longitudinal strigae in posterior half, with abundant fine setae as follows: anterior margin with row of ~10 setae, medial pair cruciate, lateral to these an upright pair, and lateral to those three pairs medioclinate setae; lateral margins with row ~8 erect setae of variable lengths; dorsal surface with two lateral rows of five short setae each; a short, anteromedial, cruciate pair, and longer posteromedial pair (Fig. 3). Wing membranes micronodulose and with numerous minute setae. LC 1 length 0.28 mm, width at level of medial lobe 0.24 mm; LC 2 length 0.36 mm; RC 1 length 0.26 mm, RC 2 length 0.35 mm. Female: Unknown. Etymology. The specific epithet is a patronym honoring Mr. Ailan Hurley-Echevarria for his diligent efforts in processing and screening amber, during which he personally found one of the two specimens.

Discussion
The phylogeny of most webspinner lineages remain contentious and in a state of flux. More importantly, numerous undescribed genera and species are known in collections and will likely have a strong influence on any estimations of relationship. It is therefore challenging to make fine determinations of the closest relatives for the Cambay amber fossils. Kumarembia can be placed within the Archembioidea clade by the 10T with a membranous area occupying the base and center of the sclerite, 10R and 10L connected by a thin basal bar, and 10RP 2 present, short, and thumb-like. The genus can be placed within the Scelembiidae [= Group C of Szumik (2004); Group A = Archembiidae s.str., Group B = Pachylembiidae] by the rectangular and centered HP and the shape of 10LP 1 which is a curved process and apically forked (simple, curved, and externally laminate in Pachylembiidae). This is quite significant given that other members of the clade occur in sub-Saharan Africa (Angola, Congo, Tanzania, Uganda) or in South America, particularly southern South America (e.g., Argentina, Bolivia, Brazil, and Peru, although Pararhagadochir is more widespread, extending as far north as Colombia and Venezuela). Accordingly, the discovery of a scelembiid in Cambay amber appears to represent one of the only Gondwanan elements of the fauna, while most other taxa show considerably different biogeographic affinities (Rust et al. 2010). As noted, relationships within Embiodea are contentious, with considerable cladistic inquiry revising phylogenetic hypotheses (e.g., Szumik 2004;Szumik et al. 2008;Klass and Ulbricht 2009). As these hypotheses of relationship continue to stabilize it will be interesting to explore further and refine the specific affinity of K. cambayensis with particular clades within Scelembiidae.