Phylogeny of the subfamily Larentiinae (Lepidoptera: Geometridae): integrating molecular data and traditional classifications

Larentiinae are the second largest subfamily of Geometridae, with more than 6200 described species. Despite recent advances in molecular systematics of geometrid moths, phylogenetic relationships between the numerous subgroups of Larentiinae are poorly known. In this study we present the most comprehensive attempt to date to resolve the phylogeny of Larentiinae, having sampled at least one species from all currently recognized 23 tribes. Fragments of one mitochondrial (COI) and eight nuclear (EF‐1α, WGL, GAPDH, RPS5, IDH, MDH, CAD and 28S) genes were sequenced, for a total of 6939 bp. Maximum likelihood and Bayesian analyses resulted in identical well‐resolved phylogenetic trees, which had maximum or near‐maximum support values at most nodes. Almost all conventionally recognized tribes represented by more than one genus were found to be monophyletic. Close to the root of Larentiinae, six tribes branch off the main lineage one after another, with Dyspteridini being sister to all other members of the subfamily. The rest of larentiines are divided into two very diverse lineages, comprising eight and at least ten tribes, respectively. There were just three findings incongruent with the conventional tribal subdivision of the subfamily. First, the genera Collix Guenée and Anticollix Prout formed a separate, previously unrecognized but well‐supported clade at the tribe level. Second, the Palaearctic genus Pelurga Hübner was placed apart from Larentia Treitschke and Mesoleuca Hübner, which were the other members of Larentiini in this analysis. Third, Cataclysmini appeared together with genera belonging to Xanthorhoini, leaving the latter paraphyletic. The Neotropic genus Oligopleura Herrich‐Schäffer is shown to belong to the tribe Euphyiini ( comb.n. ) according to both molecular data and male genital morphology. The results and the tribal classification of Larentiinae are discussed with reference to the principal publications since the end of the 19th Century. We conclude that the current tribal classification of Larentiinae is not controversial from the phylogenetic point of view and that its increasing complexity has merely reflected the accumulation of information, mainly through different methods of biosystematic study having become available for researchers. Our results indicate that diurnal lifestyle, accompanied by conspicuous coloration, has evolved independently in several subgroups of Larentiinae.