TreeBASE Search-Matrices for study 95

@INCOLLECTION{TreeBASE2Ref15280, author = {Michael J Donoghue and J. A. Doyle}, title = {Phylogenetic analysis of angiosperms and the relationships of Hamamelidae.}, year = {1989}, keywords = {}, doi = {}, url = {}, pmid = {}, booktitle = {Evolution, Systematics, and Fossil History of the Hamamelidae}, isbn = {}, publisher = {}, address = {Oxford, United Kingdom}, editor = {Peter R. Crane and S. Blackmore}, pages = {17--45}, abstract = {In order to explore the basal radiation of angiosperms and the phylogenetic position of 'Hamamelidae', we conducted a numerical cladistic analysis of monosulcate taxa ('Magnoliidae', monocots) and six tricolpate groups (assumed to be appropriate placeholders for 'Rosidae', 'Amentiferae', and other 'higher dicots'). Based on outgroup comparison with Bennettitales plus Gnetales and with Mesozoic seed ferns, angiosperms are best rooted either within Magnoliales or with Magnoliales forming a basal clade. The remaining angiosperms are united by columellar exine structure. Within this group there are three main clades: i) Laurales, including Austrobaileya, Calycanthaceae, Trimeniaceae, and Chloranthaceae; ii) Winteraceae plus Illiciales; and iii) a group united by palmate leaf venation and differentiated stamen filaments, which contains the tricolpate groups (except Illiciales) and a 'paleoherb' clade consisting of Lactoris, Aristolochiaceae, Piperaceae plus Saururaceae, Nymphaeaceae plus Cabombaceae, and monocots. Within the palmate clade, Trochodendrales plus Hamamelidales and Ranunculidae plus Nelumbo may form a clade united by tricolpate pollen and loss of ethereal oil cells, or Ranunculidae and Nelumbo may be linked with the paleoherbs based on anomocytic stomata and herbaceous habit. Under both arrangements the chloranthoid teeth of Chloranthaceae are not homologous with those in Hamamelidae and Ranunculidae, but there are almost equally parsimonious trees in which they may be homologous. Studies of Early Cretaceous cordate leaves and tricolpate pollen and of living paleoherb groups could provide valuable tests of these results. In future cladistic studies of Hamamelidae and other 'higher' dicots, Ranunculidae and paleoherbs should be considered closer outgroups than Magnoliales or Chloranthaceae.} }