TreeBASE Search-Trees for study 10429

@ARTICLE{TreeBASE2Ref18890, author = {John Jeffrey Schenk and Larry Hufford}, title = {Phylogeny and Taxonomy of Mentzelia Section Bartonia (Loasaceae)}, year = {2011}, keywords = {Bayes factors , hypothesis test , plant systematics , polyphyletic species , Shimodaira-Hasegawa test}, doi = {10.1600/036364411X583673}, url = {http://}, pmid = {}, journal = {Systematic Botany}, volume = {36}, number = {3}, pages = {711--720}, abstract = {Abstract? Taxonomic problems in Mentzelia section Bartonia (Loasaceae) are addressed using phylogeny reconstructions based on nuclear ribosomal DNA sequences from the ITS and ETS regions. Our results indicate sect. Bartonia is monophyletic and consists of two wellsupported, species-rich clades. One of these two deepest clades consists of the Great Plains M. decapetala and a group of species centered in the North American intermountain region that have been described as subshrubby; whereas, the second deepest clade is more widespread and includes taxa from the Chihuahuan and Sonoran deserts as well as the Great Plains, Rocky Mountains, and intermountain region. Hypothesis tests applying the Shimodaira-Hasegawa (SH) test and Bayes factors (BF) rejected unequivocally the monophyly of (1) the ?subshrubby? group, suggesting multiple origins of the ?subshrubby? form; (2) M . multicaulis s. l., which consists of disparate clades we propose as separate species; and (3) M . multiflora s. l., which was recovered as highly polyphyletic. Hypothesis tests were equivocal, however, in regard to the monophyly of (1) M . marginata , M. paradoxensis , and M . cronquistii ; (2) M . oreophila s. l.; and (3) M. pumila s. l. We suggest following narrow taxonomic approaches to the circumscriptions of M . multicaulis , M . multiflora , and M. pumila and advocate further studies of M . oreophila and the complex including of M . marginata , M. paradoxensis , and M . cronquistii . Our results provide the most comprehensive phylogenetic hypothesis of sect. Bartonia to date, however, more variable markers will be needed to resolve a well supported phylogeny.} }