This Ph.D project began with an existing phylogeny of the dorid nudibranch genus Halgerda (Mollusca: Gastropoda: Opisthobranchia) derived from a traditional morphological character data set. In this first evolutionary model, I used the assumption of maximum parsimony and wished to further test this model using additional biological information and additional sources of data. For the present study, I augmented the original data matrix with a prototypical character set that includes sperm ultrastructure, natural products chemistry and DNA sequence (cytochrome oxidase subunit I, COI) data. I then apply various phylogenetic analytical techniques using this expanded data matrix to determine if the new characters can be used to increase the robustness of my original, morphologically-based phylogeny.
This thesis is multi-disciplinary and has three main components, morphological (ultrastructure of sperm), molecular (DNA sequence data) and chemical (presence or absence of chemical natural products and their source). Thus, the chapters reflect this organization. In Chapter 1, the Introduction, first I describe general concepts such as the reasons for doing a cladistic analysis of a genus. I also provide background on the practice of combining character data sets to assess the phylogenetic relationships of a group of organisms. I discuss reasons for differences in the trees obtained for various data sets. Chapter 1 also provides the general background for each of the three main components of the thesis. Additional specific background is provided at the introduction of each chapter.
In Chapter 2, the DNA sequence data is presented for the COI gene of thirteen species of Halgerda. This is the first time this gene has been sequenced for this group of molluscs and also the first time a gene tree is compared to a species tree within the Discodorididae. The cladistic analyses using COI sequence data for available species shows a similar topology to the trees obtained using morphological characters for the same species. When the gene tree is compared to the tree obtained by placing all described Halgerda species into the analysis the general topology is retained. That is, the most basal species remain the most basal and the most derived species remain so. Some rearrangement is noted in the terminal taxa. The study concludes that COI is a valuable tool for phylogenetic analysis at the species level within the Nudibranchia.
Chapter 3 covers the specifics of Halgerda sperm ultrastructure and includes the comparison to previous work published on Halgerda. Halgerda is placed in the Discodorididae of the Cryptobranchia by contemporary taxonomists who use phylogenetics to propose evolutionary hypotheses. The recent classification of the cryptobranch dorids (Cryptobranchia) as a monophyletic taxon is used as a basis for comparison. Thus, Chapter 3 includes descriptions of other genera in the same Family, along with genera from a more basal Family. The other group included in my study is the Chromodoridae. The results of this part of the research demonstrate consistencies between Halgerda sperm morphology and other taxa within the Cryptobranchia, but no characters that unite Halgerda. The selection of outgroups for this thesis is also shown to be appropriate based on sperm morphology.
Chapter 4 presents for the first time, the natural products chemistry of Asteronotus cespitosus, the sister taxon to Halgerda. Both genera are within the Discodorididae, a group for which only a small number of genera have been examined for natural products. The present study shows that compounds found in Asteronotus vary geographically, are selectively sequestered and are cytotoxic to murine leukemia cells. This represents the first time the characteristic halogenated metabolites that typify the sponge Dysidea herbacea have been reported from a carnivorous mollusc, which implies a dietary derived origin as opposed to de novo synthesis.
Chapter 5 describes the natural products chemistry from the extraction of two species of Halgerda. The specimens examined are from the east and west coasts of Australia. Although only small amounts of material were available for this study, some interesting results are already apparent. Chemical differences are noted between the two closely related genera Asteronotus and Halgerda. In fact, Halgerda contains compounds that are much more complex that those found in Asteronotus. No geographic comparison was possible for the species of Halgerda examined during this study, as insufficient material was available for detailed chemical analysis of the specimens. Implications of the findings are presented. One conclusion from this part of the research is that species of Halgerda do not share similar sponge food prey. Different chemical compounds are indicated by the diverse results that came from the analyses. Comparisons between the nuclear magnetic resonance spectra, the gas chromatography profiles and the liquid chromatography-mass spectrometry spectra are quite different between the two species of Halgerda analyzed for this study.
Finally, in Chapter 6, the results of combining the diverse data collected are presented along with the final conclusions of my research. The resolution of the trichotomies apparent in the original tree derived from the morphological data set is presented. The further support gained for the major clades of Halgerda are discussed. The relationship of Halgerda to the Discodorididae is considered and the final consistency and retention indices for the most parsimonious phylogeny of the study group are reported. A matrix is presented that encapsulates the four evaluating criteria addressed in this thesis. The criteria are cost, investment in time, efficiency (gaining a reliable test or measure) and robustness (strength of support). The conclusion from this matrix is that for a species-level phylogenetic study of the Nudibranchia, morphological and molecular data appear to offer the greatest amount of information and resolution. Sperm ultrastructure and natural products chemistry are useful at higher taxonomic classification levels.