The population dynamics of five remnant rock-wallaby populations (Petrogale lateralis) persisting on granite outcrops in the central wheatbelt region of Western Australia were monitored over a six year period. From 1979 to 1982 all populations remained relatively static or declined for unknown reasons, but circumstantial evidence implicated fox predation. A fox control program was implemented in 1982 on two outcrops and was maintained for four years with the result that the two resident rock-wallaby populations increased by 138 and 223%. Two rock-wallaby populations occupying sites not subjected to fox control declined by 14 and 85%, and the third population increased by 29%. It was concluded that the fox has probably been a significant factor in the demise and decline of native mammals in the past, and that surviving populations are still at risk. Control of predation pressure on nature reserves was shown to be feasible from a management perspective.

No abstract

The population dynamics of five remnant rock-wallaby populations (Petrogale lateralis) persisting on granite outcrops in the central wheatbelt region of Western Australia were monitored over a six year period. From 1979 to 1982 all populations remained relatively static or declined for unknown reasons, but circumstantial evidence implicated fox predation. A fox control program was implemented in 1982 on two outcrops and was maintained for four years with the result that the two resident rock-wallaby populations increased by 138 and 223%. Two rock-wallaby populations occupying sites not subjected to fox control declined by 14 and 85%, and the third population increased by 29%. It was concluded that the fox has probably been a significant factor in the demise and decline of native mammals in the past, and that surviving populations are still at risk. Control of predation pressure on nature reserves was shown to be feasible from a management perspective.

The ability to confidently identify or exclude a population as the source of an individual has numerous powerful applications in molecular ecology. Several alternative assignment methods have recently been developed and are yet to be fully evaluated with empirical data. In this study we tested the efficacy of different assignment methods by using a translocated rock-wallaby (Petrogale lateralis) population, of known provenance. Specimens from the translocated population (n = 43), its known source population (n = 30) and four other nearby populations (n = 19-32) were genotyped for 11 polymorphic microsatellite loci. The results identified Bayesian clustering, frequency and Bayesian methods as the most consistent and accurate, correctly assigning 93-100% of individuals up to a significance threshold of P = 0.01. Performance was variable among the distance-based methods, with the Cavalli-Sforza and Edwards chord distance performing best, whereas Goldstein et al.'s (deltamu)2 consistently performed poorly. Using Bayesian clustering, frequency and Bayesian methods we then attempted to determine the source of rock-wallabies which have recently recolonized an outcrop (Gardners) 8 km from the nearest rock-wallaby population. Results indicate that the population at Gardners originated via a recent dispersal event from the eastern end of Mt. Caroline. This is only the second published record of dispersal by rock-wallabies between habitat patches and is the longest movement recorded to date. Molecular techniques and methods of analysis are now available to allow detailed studies of dispersal in rock-wallabies and should also be possible for many other taxa.

No abstract