Systematics and Biogeography of the Clusioid Clade (Malpighiales)

The clusioids are a clade of flowering plants in the diverse rosid order Malpighiales. It includes five families (i.e., Bonnetiaceae, Calophyllaceae, Clusiaceae sensu stricto, Hypericaceae, and Podostemaceae) that form a conspicuous element of tropical forests worldwide and are economically important. Their phylogenetic and biogeographical history has remained uncertain, however, which has hindered our understanding of their evolution. I conducted the first taxon-rich multigene analysis of this important clade to clarify their phylogenetic relationships (Chapter 1). Plastid (cp:matK, ndhF, and rbcL) and mitochondrial (mt: matR) nucleotide sequence data from nearly 200 taxa produced a well-resolved clusioid phylogeny and indicate that several traditionally recognized genera are not monophyletic. These results provide a strong framework for improving the classification of the group. To further determine the placement of several key taxa lacking molecular data, especially the ancient fossil rosid Paleoclusia (-90 Myr), I assembled a morphological data set that I analyzed in combination with the cp and mt data (Chapter 2). My results support previous hypotheses of phylogenetic relationships for extant taxa and indicate that Paleoclusia is weakly placed as a member of the clusioid subclade Clusiaceae sensu stricto. Finally, I inferred molecular divergence estimates and ancestral ranges for the clade to test the hypothesis iii that the pantropical distribution of many clusioid subclades is attributable to ancient Gondwanan vicariance (Chapter 3). The clusioids are ideal for examining this topic due to their well-sampled and strongly supported phylogeny, pantropical distribution, and ancient fossil record. Our results suggest a single Gondwanan vicariant event early in the history of the clade, followed by prevalent dispersal throughout the Cenozoic, most of which occurred after the mid-Eocene. These results are consistent with a growing body of literature that suggests that many traditionally recognized angiosperm clades are far too young for their distributions to have been influenced strictly by Gondwanan vicariance. Instead, it appears that dispersal is a more likely explanation for many Gondwanan distributions in angiosperms including the clusioids