These could be new mutations that occurred during the exponential growth of the KI population. On the assumption of uninterrupted exponential growth, the population growth rate on KI can be assessed, for it is known that the population went from N 0 = 18 at the time of the founding to N t = 27 000 in 2004, 80 years later. If we assume a koala generation time of 5 years (Martin and Handasyde 1999), the number of generations is 16. Because N t = N 0 e rt , then r = [ln (N t /N 0 )]/t = 0.4570, equivalent to a per-generation rate of increase, l, of 1.5794. This value can be used to calculate the size of the population at each intervening generation, and thus the total number of individuals that have been available for mutation. This gives an estimate of 46 546 individuals in the pedigree leading to the present population. The mutation rate on KI can be derived from the following formula: mutation number divided by (microsatellite loci ¾ individuals) = 4/(15 ¾ 46 546) = 5.7 ¾ 10 -6 . This number is slightly lower than the range of the microsatellite mutation rates for eutherian mammals (between 10 -3 and 10 -5 ; Dallas 1992; Banchs et al. 1994;Ellegren 1995), possibly because of either the small number of koalas genotyped, or the assumptions inherent in the calculations.Abstract. Habitat destruction and fragmentation, interactions with introduced species or the relocation of animals to form new populations for conservation purposes may result in a multiplication of population bottlenecks. Examples are the translocations of koalas to French Island and its derivative Kangaroo Island population, with both populations established as insurance policies against koala extinction. In terms of population size, these conservation programs were success stories. However, the genetic story could be different. We conducted a genetic investigation of French and Kangaroo Island koalas by using 15 microsatellite markers, 11 of which are described here for the first time. The results confirm very low genetic diversity. French Island koalas have 3.8 alleles per locus and Kangaroo Island koalas 2.4. The present study found a 19% incidence of testicular abnormality in Kangaroo Island animals. Internal relatedness, an individual inbreeding coefficient, was not significantly different in koalas with testicular abnormalities from that in other males, suggesting the condition is not related to recent inbreeding. It could instead result from an unfortunate selection of founder individuals carrying alleles for testicular abnormalities, followed by a subsequent increase in these alleles' frequencies through genetic drift and small population-related inefficiency of selection. Given the low diversity and possible high prevalence of deleterious alleles, the genetic viability of the population remains uncertain, despite its exponential growth so far. This stands as a warning to other introductions for conservation reasons.

No abstract