Population genetics, ecology and systematics of Indo-Australian scombrid fishes, with particular reference to skipjack tuna (Katsuwonus pelamis)

Abstract

This thesis is essentially concerned with the use of electrophoretically-detected genetic variation, in combination with mark-recapture data, to describe the genetic structure of a high vagility species, the skipjack tuna (Kateuwonus pelamis), in the Info-Australian region. Implications for management of the resource are then considered, and the results related where possible to other members of the family Scombridae. With a favourble return rate (7.6%) from the 9547 releases of tagged skipjack, it has been possible to discern cyclical movement patterns. These appear to be related to large scale oceanographic events leading to enhanced productivity. An hypothesis to explain these movements recognizes resident (island associated) and nomadic componenets of the resource. On the basis of these and subsequent tagging experiments, dispersal of adults is quite limited, with < 1% of nett displacements exceeding 1000 nautical miles. Providing dispersal of planktonic phases is of similar scope, the potential for gene flow within one generation is considerably less than in some other scombrids. Although hampered to some extent by the reliance on variation at one enzyme locus (serum esterase, ESJ) the electrophoretic studies produced results compatible with the mark-recapture data. The main feature of this variation was a cline in E1SJ frequencies across 12,000 km, matching the longitudinal extent of both the known spawning areas and the distribution of islands. It therefore appears likely that open ocean-island interactions play an important role in skipjack ecology, and in selective action on the ESJ locus. The relative roles of selection and migration in maintaining the cline remain unclear but an isolation-by-distance model may provide the best fit to the available data. A notable feature of the cline is that variance at any point is considerably greater than expected by chance. Time-series sampling at a site in Papua New Guinea has shown that this is independent of fish size and other factors and is related in part to episodic influzes of groups of skipjack with atypical gene frequencies. The complexity and continuity of recruitment into the study area did not allow particular cohorts to be monitored genetically. Heterozygosities at 26 loci in nineteen Indo-Australian scombrids showed considerable inter-locus and inter-specific variation, with maximum values observed in large highly mobile, widely distributed species. The data do not land themselves to critical test of neutralist and selectionist hypoteses but have been useful in demonstrating that most scombrids harbour more markers on which to base genetic studies than was the case for skipjack. The electrophoretic data were able to clarify the taxonomic status of two Grammatorcynus morphs, which are clearly good species, and were also used for phenetic and cladistic analyses of interspecific relationships. These showed good agreement with existing schemes based on morphological characters, but indicated that major groups within the Scombridae have long been separated and interpretations of scombrid evolution which regard morpho-physiological specializations as sequential may need some reappraisal. Zoogeographical studies appear likely to benefit from insights provided by electrophoretic comparisons.