Phylogeny of Membracoidea (Hemiptera: Auchenorrhyncha) based on transcriptome data

Abstract The suborder Auchenorrhyncha (“true hoppers”) comprises nearly half of known Hemiptera, with >43,000 known species of sap‐sucking herbivores distributed worldwide, including many important agricultural pests and vectors of plant disease. More than half of the known Auchenorrhyncha belong to superfamily Membracoidea (leaf‐ and treehoppers), which has been a source of phylogenetic contention for many years. To construct an improved backbone phylogeny of this superfamily, we obtained transcriptome data for multiple representatives of all 5 previously established extant families and nearly all subfamilies to test their monophyly and relationships. 138 taxa (132 Membracoidea and 6 outgroups) were sampled with an emphasis on families Cicadellidae and Membracidae, which were paraphyletic as previously defined by most authors, several problematic subfamilies (Aphrodinae, Eurymelinae, Ledrinae, Nicomiinae, Stegaspidinae and Tartessinae). We analysed different combinations of data sets (amino acid, complete nucleotide and degeneracy‐coded nucleotide) using different modelling schemes. The resultant trees based on different analyses are congruent in most nodes. Discordant nodes mainly pertain to relationships among cicadellid subfamilies and tribal relationships within Aphrodinae and Eurymelinae. Analyses of gene‐ and site concordance factors and quartet scores indicate that this instability is largely attributable to an overall lack of informative characters across genes and sites rather than strongly supported conflict among genes. According to the congruent nodes, we make the following revisions: combine Stegaspidinae and Centrotinae into a single subfamily, Centrotinae sensu lato; restore Stenocotini from Tartessinae to its original position in the Ledrinae; and transform Holdgatiella Evans from Nicomiinae to Melizoderinae. In addition, to solve the paraphyly of both Cicadellidae and Membracidae, a preferred option would be to combine all five previously recognized families into a single family, Membracidae sensu lato; the other option could be to render Cicadellidae monophyletic by excluding Megophthalminae and Ulopinae from Cicadellidae and elevating them to status as separate families.