A phylogenetic contribution to the study of the evolution and historical biogeography of demosponges (phylum Porifera)

The class Demospongiae contains phylogenetically, and ecologically diverse lineages that are represented in nearly every marine and freshwater habitat globally. Despite being an intellectually attractive study group to diverse disciplines, many basic questions about the general ecology, biogeography, and systematics of demosponges remain unanswered due primarily to a paucity of morphological characters that are useful for distinguishing species boundaries and lineage relationships. The primary objective of this dissertation is to address the evolution, biogeography, and systematics of demosponge lineages at diverse levels of phylogeny. Chapter 1 argues the importance of establishing a robust demosponge phylogenetic hypothesis for studies of both sponge biology and evolution, and metazoan phylogeny and early animal evolution. Furthermore, an efficient strategy is outlined to approach the problem of demosponge relationships and potential epistemological obstacles are identified. Chapter 2 uses 28S rDNA (LSU) and Cytochrome Oxidase Subunit I (COI) data from 109 demosponge species to test hypotheses of relationship within and between 'orders', to test spicular and reproductive character homology, and to obtain minimum age estimates for major divergences within the clade. These data are expanded in Chapter 3 to refine the relationships between members of the order Hadromerida and the order Poecilosclerida using a longer fragment of the LSU from 73 demosponge taxa. Finally, Chapter 4 uses sequence data from the rapidly evolving nuclear Internal Transcribed Spacer Regions (ITS) to characterize species boundaries and biogeographic relationships of Placospongia---a taxon that is morphologically homogeneous within and between distant localities. The general conclusions of these chapters follow: (1) demosponge diversity is potentially underrepresented at all levels by the existing classification and exemplar coding strategies for sponge higher taxa are not yet justified; (2) LSU data support hypotheses of relationship that indicate high levels of morphological homoplasy within the class Demospongiae that contains lineages as old as other animal phyla; (3) as molecular sequence data are obtained for more demosponge lineages many of the long-held ideas about sponge phylogeny, biogeography, ecology, and character homology are likely to be replaced or modified. Current ideas about sponge biology are on the threshold of advancement and change.