To investigate vascular-plant phylogeny at deep levels of relationship, I collected and analyzed a large set of plastid-DNA data comprising multiple protein-coding genes and associated noncoding regions. I addressed questions relating to overall tracheophyte phylogeny, including relationships among the five living lineages of seed plants, and within two of the largest living gymnosperm clades (conifers and cycads). I also examined relationships within and among the major lineages of monilophytes (ferns and relatives), including their relationship to the remaining vascular plants. Overall, I recovered three well-supported lineages of vascular plants: lycophytes, monilophytes, and seed plants. I inferred strong support for most of the phylogenetic backbones of cycads and conifers. My results suggest that the cycad family Stangeriaceae (Stangeria and Bowenia) is not monophyletic, and that Stangeria is instead more closely related to Zamia and Ceratozamia. Within the conifers, I found Pinaceae to be the sister-group of all other conifers, and I argue that two conifer genera, Cephalotaxus and Phyllocladus (often treated as monogeneric families) should be recognized under Taxaceae and Podocarpaceae, respectively. Systematic error likely affects inference of the placement of Gnetales within seed-plant phylogeny. As a result, the question of the relationships among the five living seed-plant groups still remains largely unresolved, even though removal of the most rapidly evolving characters appears to reduce systematic error. Phylogenetic analyses that included these rapidly evolving characters often led to the misinference of the “Gnetales-sister” hypothesis (Gnetales as the sister-group of all other seed plants), especially when maximum parsimony was the inference method. Filtering of rapidly evolving characters had little effect on inference of higher-order relationships within conifers and monilophytes, and generally resulted in reduced support for backbone relationships. Within the monilophytes, I found strong support for the majority of relationships along the backbone. These were generally congruent with other recent studies. Equisetaceae and Marattiaceae may be, respectively, the sister-groups of the remaining monilophytes and of the leptosporangiate ferns, but relationships among the major monilophyte lineages are sensitive to the outgroups used, and to long branches in lycophytes.

To investigate vascular-plant phylogeny at deep levels of relationship, I collected and analyzed a large set of plastid-DNA data comprising multiple protein-coding genes and associated noncoding regions. I addressed questions relating to overall tracheophyte phylogeny, including relationships among the five living lineages of seed plants, and within two of the largest living gymnosperm clades (conifers and cycads). I also examined relationships within and among the major lineages of monilophytes (ferns and relatives), including their relationship to the remaining vascular plants. Overall, I recovered three well-supported lineages of vascular plants: lycophytes, monilophytes, and seed plants. I inferred strong support for most of the phylogenetic backbones of cycads and conifers. My results suggest that the cycad family Stangeriaceae (Stangeria and Bowenia) is not monophyletic, and that Stangeria is instead more closely related to Zamia and Ceratozamia. Within the conifers, I found Pinaceae to be the sister-group of all other conifers, and I argue that two conifer genera, Cephalotaxus and Phyllocladus (often treated as monogeneric families) should be recognized under Taxaceae and Podocarpaceae, respectively. Systematic error likely affects inference of the placement of Gnetales within seed-plant phylogeny. As a result, the question of the relationships among the five living seed-plant groups still remains largely unresolved, even though removal of the most rapidly evolving characters appears to reduce systematic error. Phylogenetic analyses that included these rapidly evolving characters often led to the misinference of the “Gnetales-sister” hypothesis (Gnetales as the sister-group of all other seed plants), especially when maximum parsimony was the inference method. Filtering of rapidly evolving characters had little effect on inference of higher-order relationships within conifers and monilophytes, and generally resulted in reduced support for backbone relationships. Within the monilophytes, I found strong support for the majority of relationships along the backbone. These were generally congruent with other recent studies. Equisetaceae and Marattiaceae may be, respectively, the sister-groups of the remaining monilophytes and of the leptosporangiate ferns, but relationships among the major monilophyte lineages are sensitive to the outgroups used, and to long branches in lycophytes.

To investigate vascular-plant phylogeny at deep levels of relationship, I collected and analyzed a large set of plastid-DNA data comprising multiple protein-coding genes and associated noncoding regions. I addressed questions relating to overall tracheophyte phylogeny, including relationships among the five living lineages of seed plants, and within two of the largest living gymnosperm clades (conifers and cycads). I also examined relationships within and among the major lineages of monilophytes (ferns and relatives), including their relationship to the remaining vascular plants. Overall, I recovered three well-supported lineages of vascular plants: lycophytes, monilophytes, and seed plants. I inferred strong support for most of the phylogenetic backbones of cycads and conifers. My results suggest that the cycad family Stangeriaceae (Stangeria and Bowenia) is not monophyletic, and that Stangeria is instead more closely related to Zamia and Ceratozamia. Within the conifers, I found Pinaceae to be the sister-group of all other conifers, and I argue that two conifer genera, Cephalotaxus and Phyllocladus (often treated as monogeneric families) should be recognized under Taxaceae and Podocarpaceae, respectively. Systematic error likely affects inference of the placement of Gnetales within seed-plant phylogeny. As a result, the question of the relationships among the five living seed-plant groups still remains largely unresolved, even though removal of the most rapidly evolving characters appears to reduce systematic error. Phylogenetic analyses that included these rapidly evolving characters often led to the misinference of the “Gnetales-sister” hypothesis (Gnetales as the sister-group of all other seed plants), especially when maximum parsimony was the inference method. Filtering of rapidly evolving characters had little effect on inference of higher-order relationships within conifers and monilophytes, and generally resulted in reduced support for backbone relationships. Within the monilophytes, I found strong support for the majority of relationships along the backbone. These were generally congruent with other recent studies. Equisetaceae and Marattiaceae may be, respectively, the sister-groups of the remaining monilophytes and of the leptosporangiate ferns, but relationships among the major monilophyte lineages are sensitive to the outgroups used, and to long branches in lycophytes. / Science, Faculty of / Botany, Department of / Graduate