Abstract
This thesis was initiated to address two of the questions posed by Stewart's 1995 study on the biology of Astrobrachion constrictum; Is gene flow between fiords restricted in this ophiuroid species and are the colour morphs as different genetically as they are visually? Specimens of A. constrictum were collected from four sites
within Doubtful Sound and one site in each of Nancy Sound, Preservation Inlet and Chalky Inlet. The genetics of these populations of the snake star were assessed using mitochondrial DNA and allozyme analysis. These results were analysed using G-test and AMOV A. Both G-test and AMOV A anlaysis showed that there was no significant genetic differentiation between populations within the same fiords (G[21] = 11.97; P > 0.9) (Fsc = -0.037, P > 0.05) or among fiords (G[21] = 23.32; P > 0.2) (FST = 0.013, P > 0.05). This result was unexpected as the Fiordland environment appears to present several barriers to dispersal, and genetic differentiation of Fiordland populations has been demonstrated in other echinoderm species. Genetic differences between the five colour morphs of A. constrictum were also assessed using mitochondrial DNA and allozyme analysis. G-test and AMOV A analysis of these results also showed no significant differences between colour morphs (G[28] = 13.88; P > 0.9) (F[ST] = -0.089, P > 0.05) indicating that they are conspecific. While studying the population genetics of the snake star two individuals were discovered which appeared to be heteroplasmic. Through the use of PCR cloning, SSCP and sequencing, the presence of heteroplasmy within these two individuals was demonstrated. This is the first reported case of heteroplasmy in an ophiuroid species and probably arose through paternal leakage. Cytochrome Oxidase I sequence obtained from the population analysis was also used in conjunction with sequences from Genbank to assess the phylogeny of echinoderms. Several studies have addressed echinoderm phylogeny over the past century but have failed to clarify the relationship between the echinozoa ( echinoids and holothuroids ), ophiuroids and asteroids. A relatively fast evolving gene was used in this study in an attempt to clarify this relationship. Of the nineteen trees generated with PAUP (Swofford, 1993), only one gave a topology indicating monophyly for all five classes. This was trimmed and compared with three previous hypotheses using the Kishino-Hasegawa test. Results from this showed no significant difference between hypotheses. This is probably due the large amount of noise introduced into the study through the use of a rapidly evolving gene.