Cat ownership in Australia is declining compared to an increasing trend of cat ownership in the United Kingdom, United States and Europe. The decline in Australia may be linked to concerns over perceived impacts of cat predation and an associated dislike of cats. However, while there are numerous studies on feral cats and their impacts on declining native fauna, the impact of pet cats on suburban wildlife or fauna in remnant bushland is relatively unknown although there is a wide perception of risk. The primary aim of this thesis was to apply the precautionary principle to the question of the putative impact of pet cats on the abundance and diversity of small mammals in urban bushland adjacent to low-density suburbia in the City of Armadale, a municipality on the south-east fringe of Perth, Western Australia. At the time of writing, Western Australia is yet to introduce state legislation governing cat control although many local councils within the state have either implemented or are in the process of implementing cat regulations. The precautionary principle was deemed an ideal approach to this question, because it provides a rationale for deciding on possible actions where both the potential risk to environmental values and the uncertainty about possible impacts are high. In such cases the precautionary principle requires two broad lines of action: firstly, detailed consultation with stakeholders to determine their perceptions of risk and the actions they are prepared to take to reduce it and, secondly, research to reduce uncertainty. With regard to stakeholder consultation, local residents were surveyed in regard to their attitudes and current cat husbandry practices. A substantial proportion of respondents within this municipality believed cat regulations were necessary (75% of owners and 95% of non-owners). At least 70% of both owners and non-owners agreed with the propositions that cats not owned by licensed breeders should be desexed, local councils should restrict the maximum number of cats that can be owned on one property and that pet cats entering nature reserves are harmful to wildlife. Most (c.85%) cat owners agreed that they would license their cats if that became compulsory. Although fewer owners (c.60%) were prepared to keep their cats on their property at all times to protect wildlife, over 80% were willing to confine their cats at night if it was required. Owners seemed to be substantially motivated by the value of these measures in reducing injury to cats and facilitating the return of lost animals rather than concern over wildlife protection. Attempts to reduce uncertainty involved (i) assessing roaming patterns of pet cats to determine the sizes of appropriate buffer zones around nature reserves, and (ii) determining species diversity, species richness and abundance of small mammals in remnant bushland adjacent to sub-divisions with varying regulations governing cat husbandry. Radio tracking results to assess cat roaming patterns showed substantial variation in home range size between cats in high density suburbia (ranged between 0.01 ha – 0.64 ha) and those in low density suburbia (ranged from 0.07 ha – 2.86ha). Larger home range sizes of cats in the rural areas (up to 2.9 ha) suggest buffer zones of up to 500 metres around nature reserves are needed to exclude almost all roaming cats. The abundance and species richness of small mammals were investigated in four areas of remnant bushland. Two were adjacent to subdivisions where cat ownership was unrestricted, one next to a subdivision where cat ownership was prohibited and the remaining one next to a subdivision where compulsory night curfew and bells on pet cats were enforced. No definitive evidence of predatory impact by pet cats on the small mammals was found. Mammal species diversity was not significantly different between sites and species richness and absolute abundance were not higher in sites where cats were restricted. Vegetation comparisons showed significant differences in the structure and species composition of the vegetation between most sites and the mammal species richness and abundance appeared linked to ground cover density in the various sites. This factor, not cat restrictions, appeared to be the primary determinant of species richness, species diversity and absolute numbers of small mammals in these sites. This study in the City of Armadale has shown that the implementation of proposed cat legislation must have a “whole of ecosystem” approach, i.e. protecting identified remnant bushland containing biodiversity from threatening processes such as plant disease and inappropriate fire, especially arson, as well as possible predations from pet cats. Habitat restoration and protection may be more important conservation activities than regulation of cats. Regulation of cats can be done at differing levels of intensity and cost, bearing in mind that this community is receptive to regulation of some aspects of cat ownership. Community education on the values of cat confinement in regards to cat welfare might increase chances of compliance.