Taxonomic affinity of halictid bee fossils (Hymenoptera: Anthophila) based on geometric morphometrics analyses of wing shape

Abstract

Taxonomic assignment of fossils can be difficult because descriptions and phylogenetic analyses are often based on a limited number of discrete visible morphological characters. Quantitative, morphometric analyses can help to assign fossil specimens to modern groups, especially when two-dimensional features, such as insect wings, provide easily identifiable landmarks that are not likely to be deformed in the fossil specimens. Here we study taxonomic attribution of bee fossils by analysing wing shape of fossil and contemporary taxa. Our study focuses on the bee family Halictidae (Hymenoptera: Halictidae: Halictinae). Halictinae is a well-known cosmopolitan group including more than 2500 species in five tribes described from robust molecular and morphological analyses. We analysed 202 specimens of 48 species of Halictinae and Nomiinae. To analyse wing shape, we performed a Geometric Morphometrics analysis by using 19 2D-landmarks. Ordination methods, such as canonical variates analysis (CVA) and linear discriminant analysis (LDA), were used to discriminate tribes based on wing shape. CVA and LDA were both powerful enough to recover tribes previously delimited by adult morphology and DNA sequences. We then assigned fossils in CVA and LDA groups based on Mahalanobis distances. The ‘morphometric affinity’ of the two fossil taxa did not confirm their previous taxonomic attribution. By re-examining the fossil description we propose a new taxonomic attribution for Electrolictus antiquus.