Recent independent phylogenetic analyses of membracid relationships based on molecular and morphological data have identified monophyletic lineages within the family. However, the results of these studies have not fully resolved treehopper phylogeny, and relationships among some higher membracid lineages remain in doubt. Portions of three datasets (958 aligned nucleotides from elongation factor-1α, 2363 aligned nucleotides from 28S ribosomal DNA, and eighty-three morphological features of adults and nymphs) introduced in recent studies were reanalysed separately and in combination with two new molecular datasets (321 aligned nucleotides from wingless and 1829 aligned nucleotides from 18S ribosomal DNA). The results of the combined data analyses, contrary to previous analyses of morphological data alone, grouped membracids into two well-supported lineages, one comprising Stegaspidinae and Centrotinae, the other comprising Membracinae, Darninae and Smiliinae. The analyses recovered Centrotinae, Membracinae and Darninae as monophyletic groups, but Stegaspidinae was paraphyletic with respect to Centrotinae, and Smiliinae was polyphyletic with Micrutalini placed as a sister group to the clade comprising Membracinae, Darninae and Smiliinae. These results are consistent with the following hypotheses, proposed previously based on an analysis of morphological data: (1) the posterior pronotal process was derived and lost multiple times during the evolution of Membracidae; (2) Membracidae originated in the New World and reached the Old World subsequently via dispersal; (3) maternal care evolved independently multiple times and may or may not have been preceded by the acquisition of ant mutualism.