The Maintenance of Genetic Diversity in Subdivided Populations of the Mosquitofish Gambusia holbrooki
MetadataShow full item record
Sub-populations inhabiting the length of a stream may be subject to a number of substantially different spatial and temporal environmental influences. To be able to monitor the effects of localised disturbance throughout a catchment is of prime import, not only to the study of evolutionary processes, but also to the assessment of human intervention in catchment dynamics. To understand the ultimate effects of localised disturbance, either natural or man-maAe, on the evolution of sub-divided populations, it is essential to determine the patterns of genetic differentiation, the levels of genetic exchange between sub-populations, and the potential for natural selection or genetic drift associated with a particular habitat type or disturbance. The research outlined in this thesis, aimed at determining how disturbance over part of a species range interacts with gene flow to alter the evolutionary processes in each deme of a sub-divided population. These aims involved assessing the relative conthbÃ¼tion of gene flow, natural selection and genetic drift, to the genetic structure of Mosquitofish Gambusia holbrook! populations at particular times during their yearly breeding cycle. Spatial and temporal variation in Pgi allozyme frequencies were examined relative to distance, flow velocity, and bathers to dispersal, for Gambgsia hoibrooki populations inhabiting an intennitant stream in South-eastern Queensland. Isolation by distance explained only a small amount of the genetic variation between sites. Similarly, levels of differentiation could not be explained by a simple relationship between distance and flow velocity common to both catchments examined. Barriers to dispersal allowed movement only in the downstream direction during peak flooding. Different alleles occurred in high frequencies in the headwaters of each catchment. The dominance of a single allele in headwater populations is attributed to the lack of upstream movement, and the ephemeral nature of these populations. This same allele was maintained as the most common within each catchment by downstream emigration. Inter-catchment differences were therefore attributed to the random fixation of different alleles in the headwaters of each catchment, and the dispersal of these alleles downstream during peak flood events. Significant allele frequency differences between populations in one catchment occurred despite immigration. Differentiation fluctuated over a yearly cycle, and was dependent on the level of immigration entering the lower site from alternative upstream sources. Immigration from source populations was in turn related to flow velocities in the tributaries which converged above this site. All alleles were maintained in the lower reaches of both catchments despite the input of a single allele from headwater populations. Given the mechanism by which frequency differences between sites in the lower western catchment were maintained, and severe restrictions only on upstream movement, it is hypothesised that the less common alleles are preserved in the lower reaches of each catchment, by immigration originating in tributaries where alternative alleles have drifted to relatively high frequencies. Natural selection associated with saltwater inundations and the preferential movement of Ga,nbusia size classes created and maintained genetic differences between populations in the lower western catchment, not only while these populations were isolated, but also during periods of low flow. Preferential movement also enabled selection effects to be carried into adjacent populations, causing them to differentiate from populations not receiving immigrants. Selection effects were not consistent, ocurring in some years and not others. When selection took place allele frequency differences could be maintained at the site of selection for up to a year. The influence of selection at an adjacent site receiving immigrants only maintained differences between this and other sites for a period of about one month. No selection effect could compete with gene flow during high flow velocities each year. The maintenance of genetic diversity in lower reach populations was not dependent on selection associated with habitat heterogeneity in this region of the creek, but on the input of different alleles from tributaries converging upstream. Selection would only have a significant effect on the maintenance of diversity following high flow, if it had produced allele frequency differences in the upper reaches of each catchment. The restriction of upstream movement, and the small effective population sizes associated with the ephemoral nature of headwater populations, suggests genetic drift was more important in determining the final allele frequencies, and maintaining diversity in this system.
Thesis (PhD Doctorate)
Doctor of Philosophy (PhD)
Australian School of Environmental Studies