Phylogeny of Polygonia, Nymphalis and related butterflies (Lepidoptera : Nymphalidae): a total-evidence analysis

We investigated the phylogeny of butterflies in the tribe Nymphalini sensu Harvey 1991, comprising the genera Vanessa, Cynthia, Bassaris, Aglais, Inac his, Nymphalis, Polygonia, Kaniska, Antanartia, Hypanartia, Symbrenthia, My nes and Araschnia. Evidence from the mitochondrial gene ndl, the nuclear ge ne 'wingless' and from morphology/ecology/behaviour were used separately an d combined to analyse relationships. Phylogenies based on the different typ es of data agreed in many aspects of basic topology. We show that an analys is of only wing pattern characters (based on Nijhout's homology system) res ults in a topology broadly similar to the one resulting from analysis of th e complete matrix. We found support for a monophyletic Nymphalini, where Hy panartia may be the sister clade to all other genera. Mynes, Symbrenthia an d Araschnia together seem to form another basal clade. Evidence presented g ives only moderate support for a monophyletic Vanessa in the wide sense, in cluding also Cynthia and Bassaris, but strong support for the monophyly of the largely holarctic clade Aglais + Inachis + Nymphalis + Polygonia + Kani ska + Roddia. Within the latter group there is strong support for a clade c onsisting of Aglais + Inachis and for a second clade which includes Nymphal is, Polygonia (and its sister clade, the monotypic Kaniska) as well as Rodd ia E-album ( = Nymphalis vaualbum). As a consequence of this topology, Agla is is recognized as a taxon separate from Nymphalis. We present a hypothesi s of species relationships within the focal group of genera. We also analys e and discuss the implications of excluding or including ecological data in phylogenetic tree construction, when the tree is to be used for studies in phylogenetic ecology. (C) 2001 The Linnean Society of London.