Global phylogeography reveals the origin and the evolutionary history of the gypsy moth (Lepidoptera, Erebidae)

•A robust phylogeography of gypsy moth based on a comprehensive taxon sampling is proposed.•Intraspecific variation and relationship among major phylo-geographic lineages are clarified.•Discovery of a new cryptic lineage from Persian Plateau.•A new scenario for colonization and the geographic centres of origin of gypsy moths is proposed. We examined the global phylogeography of gypsy moth (Lymantria dispar L.) using molecular data based on mitochondrial and nuclear genes. Populations from all biogeographic regions of the native and introduced range of L. dispar, were sampled to fully document intraspecific and subspecies variation, identify potential cryptic species, and to clarify the relationships among major phylogeographic lineages. We recovered three major mtDNA lineages of L. dispar: Transcaucasia; East Asia + Japan; and Europe + Central Asia. The circumscription of these lineages is only partially consistent with the current taxonomic concept (i.e., L. dispar dispar; L. dispar asiatica; L. dispar japonica), with the following important discrepancies: (1) north-central Asian populations, including topotypical populations of L. dispar asiatica, may be more closely related to European rather than Asian segregates, which would require the synonymization of the taxon asiatica and establishment of a new name; (2) the Japanese populations (L. d. japonica) are not distinct from east Asian populations; (3) the presence of a distinct, unnamed mitogenomic lineage endemic to the Trancaucasus region. We demonstrated that the population from Transcaucasia contains the highest mitochondrial haplotype diversity among L. dispar, potentially indicative of an ancestral area for the entire dispar-group. Our study corroborates the endemic Hokkaido, Japan taxon Lymantria umbrosa (Butler) as the sister group to all other L. dispar populations, but the applicability of the names umbrosa versus hokkaidoensis Goldschmidt needs to be re-evaluated. The ancestral area analysis suggest that Japan was likely colonized via Sakhalin ∼1 Mya, in contrast to previous studies which have suggested colonization of the Japanese archipelago via the Korean Peninsula. Lastly, mitogenomic variation within L. dispar is incongruent with phylogenies based on nuclear DNA, as nDNA gene phylogenies did not recover the three major mtDNA lineages, and also failed to recover L. dispar and L. umbrosa as reciprocally monophyletic.