|dc.description.abstract||An increasing number of endangered species have reached the threshold prompting management decisions to commence captive breeding and reintroduction. Such intervention is costly and requires a substantial investment of resources that might otherwise be spent on alternate conservation options. It is important, therefore, that such measures make use of all available information to ensure the success of the reintroduction effort, not just in the short-term but also well into the future. The diverse range of elements to be considered include habitat factors, number and sex ratio of founders, number of populations to establish, source populations to use, timing of releases and the need to supplement the populations. All of these factors can benefit through input from molecular genetic data to improve the quality of information used in decision making. Levels of genetic variation are strongly correlated with population fitness and their potential for long-term persistence. For this thesis I examined levels of genetic diversity at neutral and functional regions of the genome for two endangered species of Australian marsupials: the western barred bandicoot, Perameles bougainville; and Queensland populations of the greater bilby, Macrotis lagotis.
These two species are under threat from similar processes: habitat destruction and modification for agriculture; predation by foxes and feral cats; changes to traditional fire regimes and competition with introduced pest species. Since European settlement, P. bougainville has become extinct on the mainland and now exists on just two islands in Shark Bay, Western Australia. Macrotis lagotis has suffered a range contraction of over 80% and the Queensland population has become disjointed from other populations to the west. Reintroduction efforts are under way for both species but, until now, the projects have not made use of molecular genetic data to inform their management decisions. I have used marker systems from nuclear microsatellite DNA, mitochondrial control region DNA and the functionally important major histocompatability complex (MHC) to assess levels of genetic diversity in natural, captive and reintroduced populations of both species. DNA was sourced from ear-tissue for P. bougainville and from ear-tissue and faecal pellets for M. lagotis.
The levels of microsatellite diversity for the two natural populations of P. bougainville (Bernier Island: HE = 0.27± 0.1, A = 1.8± 0.3; Dorre Island: HE = 0.31± 0.1, A = 2.2± 0.4) were low compared to other marsupials and significantly lower than that recorded for the natural Queensland population of M. lagotis (Astrebla Downs: HE = 0.76 ± 0.03, A = 4.31 ± 0.3). In all cases, the diversity of captive and reintroduced populations was reduced relative to their source populations except for the Dryandra captive population of P. bougainville (HE = 0.54± 0.1, A = 2.69± 0.2) which was founded following a mixed breeding strategy using individuals sourced from both natural island populations. Distribution of mtDNA haplotypes among geographical regions indicated that, for each species, populations could be combined in captive breeding programs without compromising distinct evolutionary lineages. Design of MHC assays proved difficult for M. lagotis, but for P. bougainville two separate MHC class II alleles were identified. These two alleles were fixed across all individuals in all populations suggesting that they may represent two paralogous loci in P. bougainville and that MHC diversity is unusually low for this species.
I have recommended that the recovery programs for both species be expanded to incorporate regular monitoring of molecular data to ensure that genetic diversity is retained and maximised in all populations. Where possible, the natural populations should be maintained as “pure lines” to increase overall species genetic diversity but the captive and reintroduced populations should make use of supplementary individuals from a mixture of sources to maximise variation and thus the adaptive potential of these populations in the novel environments to which they are being introduced.||