Empria parvula (Konow, 1892) *

Liston, Andrew, Prous, Marko & Macek, Jan, 2019, On Bulgarian sawflies, including a new species of Empria (Hymenoptera, Symphyta), Deutsche Entomologische Zeitschrift 66 (1), pp. 85-105 : 90-91

Empria parvula (Konow, 1892) *

Poecilosoma parvula Konow, 1892: 215. Lectotype ♂ GBIF-GISHym3784 (SDEI), here designated. Type locality: Fürstenberg in Mecklenburg, Germany, Brandenburg.

Empria pseudoklugi [pseudo-klugi sic!] Dovnar-Zapolskij, 1929: 39. Lectotype ♀ ZIN_Empria_8 (ZIN), here designated. Type locality: Sarepta, Volgograd Oblast, Russia.

Bulgarian material.

Varna: 1♂ (DEI-GISHym88775), Tsonevo 5 km S, 100 m, 42.982N, 27.451E, 06.04.2018 . 1♀, locality as previous, 08.04.2018 . 1♀ (DEI-GISHym88802), Staro Oryahovo 2 km SW, 120 m, 42.976N, 27.787E, 08.04.2018 . 1♂, locality as previous, 09.04.2018 . 1♀, 1♂, Goren Chiflik 1 km SW, 40 m, 43.001N, 27.621E, 13.04.2018 .

Empria parvula has an extensive distribution in Europe ( Taeger et al. 2006) and the Palaearctic ( Sundukov 2017; Taeger et al. 2018). According to morphological characters and genetic data, the species is very closely related to E. pravei . The only clear difference between them is the colouration of the legs in the adults. Legs in E. parvula are usually mostly black with small pale areas, but occasionally the hind tibia can be basally 2/3 whitish or yellowish. Nevertheless, the metafemur appears to be always completely or nearly completely black in E. parvula (Fig. 32 ). In E. pravei , femora are apically and tibiae basally extensively yellowish (Fig. 33 ). There could be differences also in penis valves, but because of the variation within E. parvula , the differences are not always clear (Figs 24-27 ). The valviceps seems to usually expand basally less in E. parvula than in E. pravei (Figs 24-27 ). Host plants and at least colouration of larvae are not different between E. parvula and E. pravei (Figs 34 , 35 ). Based on the sequence data of three genes that we currently have, E. parvula does not form a monophyletic group, particularly because an E. parvula specimen sampled from Bulgaria is closer to E. pravei than to other E. parvula specimens (from Estonia, Greece, and Russian Far East) (Figs 16 , 36 ). Ignoring morphological evidence, our genetic data could be interpreted as indicating either that E. pravei is a synonym of E. parvula , or that E. parvula consists of more than one species. Because in Bulgaria we found E. pravei and E. parvula in the same places at the same time and never observed overlap with regard to leg colouration, we consider E. pravei to be a distinct species. Although the existence of more than one species under the name E. parvula cannot be excluded, the data is also consistent with a single species exhibiting large genetic variation, perhaps connected with the significantly larger population size in E. parvula (distributed throughout the Palaearctic) compared to E. pravei (possibly restricted to areas not far from the Black Sea and south of the Caspian Sea). In other words, non-monophyly of E. parvula could be because of incomplete lineage sorting (maintaining of ancestral polymorphisms) due to large population size (e.g. Degnan and Rosenberg 2009). More specimens and genes of both species should be sequenced or mating experiments done to decide more reliably about species boundaries in this case.

Figures

Figure 16. Maximum likelihood tree of Empria based on three genes. Best-fit model chosen according to Bayesian information criterion was GTR + R 3. Numbers beside nodes show SH-aLRT support (%) / ultrafast bootstrap support (%) values. Support values for weakly supported branches (<90) are not shown. Letters " f " and " m " stand for " female " and " male ". Numbers at the end of the tip labels refer to the length of the sequence and the number of ambiguous positions (e. g. polymorphisms). Empria candidata and E. multicolor were used to root the tree. The scale bar shows the number of estimated substitutions per nucleotide position.

Figure 36. Maximum likelihood tree of Empria parvula and E. pravei specimens based on mitochondrial COI. Best-fit model chosen according to Bayesian information criterion was HKY + I. Numbers above branches show SH-aLRT support (%) / ultrafast bootstrap support (%) values. Support values for weakly supported branches (<90) are not shown. Letters " f " and " m " stand for " female " and " male ". Numbers at the end of the tip labels refer to the length of the sequence. The scale bar shows the number of estimated substitutions per nucleotide position.

Figures 30 - 35. Empria species. 30 Empria sexpunctata DEI-GISHym 88916, body dorsal; 31 head dorsal. 32 E. parvula DEI-GISHym 88913, ventrolateral 33 E. pravei DEI-GISHym 88735, ventrolateral. 34 E. parvula larva TUZ 615249 (Estonia) 35 E. pravei larva, reared ex ovo from DEI-GISHym 84166 (Bulgaria).

Figures 24 - 29. Penis valves of Empria. 24 Empria parvula TUZ 615399 (Estonia) 25 E. parvula DEI-GISHym 88775 (Bulgaria) 26 E. pravei DEI-GISHym 88735 (Bulgaria) 27 E. pravei DEI-GISHym 88758 (Bulgaria) 28 E. sexpunctata DEI-GISHym 88818 (Bulgaria) 29 E. aridicola paratype DEI-GISHym 80304 (Greece).