A study of the behavioural ecology of the woylie (Bettongia penicillata) was conducted between November 1986 and September 1987 on a population that had been introduced to Island A, Venus Bay in South Australia. The primary aim of the research was to establish the reasons for the success of this population of approximately 20 individuals which survives in an environment which is considerably different from the pockets of forest and woodland habitats comprising the present range of this species in Western Australia. This knowledge will assist wildlife authorities to identify suitable reintroduction sites within the former mainland range of B. penicillata. Scat analysis revealed that woylies on Island A consumed a catholic diet of dicotyledon leaf material, seasonal fruits and berries, and invertebrates. Endomycorrhizal fungal spores were common in scats during spring and are possibly derived from the gut of Scarab beetle larvae that were present in the scats at this time. Fungi are an insignificant dietary component of woylies at Venus Bay. The home ranges of woylies on Island A overlapped considerably and were significantly smaller than the "feeding" areas recorded for Western Australian conspecifics. Separate "feeding" and "nest" areas could not be identified on Island A as refuges were dispersed throughout the home range and were at a density of approximately three times that reported for Western Australian woylies. The refuge sites used by woylies on Island A were relatively simple constructions and contrasted with the more complex double-layered structures described for mainland woylies. Refuge sites on Island A were generally located beneath dense bushes. They provided little protection from prevailing temperature and relative humidity but did provide cover and ameliorate the effects of insolation. Refuges were exclusive to individuals. Males maintained a greater degree of spatial separation from one another during refuge occupation than they did from females, or females from each other. No significant difference between male and female body masses was detected, while the mean body mass recorded in December was significantly higher than at other times. The woylies at Venus Bay had significantly lower masses than those reported for a population of woylies at Tuttanning in Western Australia. The limited genetic base of the founding population and competition for food among this high density population are the most likely reasons for the lower masses of woylies on Island A. This study demonstrates that woylies can be successfully introduced into a wild, modified environment if their basic habitat requirements of suitable cover and diet are met. A broad diet, the abundance of suitable food species and adequate refuge sites within a small area, together with the absence of predators and fire, have all contributed to the success of the woylie population introduced to Island A. Active management is required to monitor and manipulate such small populations to ensure that genetic viability is maintained and the long term survival of the species is secured. Management: Recommendations 1. Monitor the population dynamics and genetic profiles of wild woylie populations in Western Australia. 2. Monitor the genetic profiles of woylies on Island A and other introduced populations. 3. Identify and acquire habitats similar to, and within, those of B. penicillata's former range, remove stock and exotic herbivores, implement and maintain predator control measures, and foster the growth of food species suitable for woylies. 4. Conduct research at proposed introduction sites to establish the fire regimes that would maintain adequate food and cover for woylies. 5. Introduce fresh genetic stock from wild Western Australian populations into the existing small populations in South Australia and simultaneously cull adults from the receiving population. 6. Maintain management practices to protect relic populations in Western Australia and newly established populations at introduction sites.