ABSTRACT Despite increasing evidence for the impact of feral cats Felis catus on native fauna in Australia, little is known of the ecology of cats, particularly factors that limit cat abundance. The ecology of the feral cat in Australia is represented by just 15 published studies on diet, only one of which has examined diet in relation to prey availability, and one study of home range behaviour. The red fox Vulpes vulpes is a significant pest to agriculture and native fauna in Australia and widespread fox removals have been proposed by the Vertebrate Biocontrol Cooperative Research Centre (VBCRC). However, there is concern that feral cats may increase compensatorily when fox populations are reduced, as has occurred in Western Australia, and therefore that predation pressure may not be alleviated on native fauna following fox control programs. This thesis is divided into two parts. First, the diet and home range size of cats is examined in relation to prey availability, and home range overlap and habitat use are determined. In the second part, several niche parameters (diet, home range and habitat use) that were potentially important resources for foxes and cats were quantified to assess the potential for competition. Avoidance and aggression between cats and foxes was examined using simultaneous radiotracking techniques and video observations. The hypothesis that foxes limit cats through interspecific competition (exploitation and interference) was then tested using a fox removal experiment. Finally, three further hypotheses were tested using a fox removal experiment to determine which factors limit feral cats at Burrendong. The four hypotheses tested were thus: i) Cats are limited independently of foxes through other factors such as food availability; ii) Foxes limit cats through interspecific competition (exploitation and/or interference); iii) Foxes limit cats through intraguild predation; iv) Cats benefit from the presence of foxes through facilitation. The diets and spatial use of feral cats were examined on agricultural land on the eastern shore of Lake Burrendong, New South Wales (32o40�S, 149o 10�E) between July 1994 and June 1997. The major land use for the area is water catchment under the agistment of sheep Ovis aries and cattle Bos taurus. The study area encompasses about 90 km2 of hilly terrain with undulating slopes that extend down to a flat foreshore area that has been extensively cleared of trees for grazing. The slopes are generally well timbered and dominated by white box Eucalyptus albens woodlands with some yellow box E. melliodora associations. Stands of cyprus pines Callitris spp. are also common. Feral cats and red foxes are established throughout the study area, and the European rabbit Oryctolagus cuniculus was abundant until the arrival of Rabbit Calicivirus Disease (RCD) in June 1996. The diet of feral cats was determined from the analysis of 499 scats. Rabbits were the staple prey of cats, with occurrence (O) in 81.6% of scats and comprising 68.4% by volume (V). Carrion (mostly eastern grey kangaroo Macropus giganteus and sheep) (O 21.5%, V 11.5%) was an important secondary food, particularly in winter and spring. Other mammalian prey included brushtail possums Trichosurus vulpecula (O 4.6%, V 2.4%), house mice Mus domesticus (O 6.2%, V 3.2%), black rats Rattus rattus (O 2.6%, V 1.4%) and a dunnart Sminthopsis sp. (probably S. murina) (O 0.2%, V 0.006%). Invertebrates (mostly Orthopterans) (O 41.5%, V 7.5%), vegetation (O 26.3%, V 3.6%), birds (O 4.2%, V 0.8%) and reptiles (O 3.4%, V 0.3%) were generally of minor importance in the diet. Few significant seasonal differences were found, although invertebrates contributed significantly less, and possums more, to the mean scat volume in winter and summer respectively. A significant dietary response was found for changes in rabbit abundance, but not for the other prey groups. Cats continued to prey heavily on rabbits after the arrival of Rabbit Calicivirus Disease, despite the relatively low numbers of rabbits. Ten months post-RCD, house mice increased in importance in the diet. However, it was not known whether this represented prey switching sensu stricto or opportunistic predation on an increased mouse population, as mouse abundance was not measured during this period. Seventy-seven cats (48 recaptures) were caught in 6762 trap nights between November 1994 and August 1996 using both cage traps and leg-hold traps. A further 18 individual cats were trapped as non-target animals by the VBCRC Fox Sterility Project and used in this study. Trapped adult cats were fitted with radio collars and their home range size, overlap and habitat use examined. Home ranges and core areas were quantified using 95% and 50% kernel utilisation distributions (KE 95 and KE 50) and minimum convex polygons (MCP 100, MCP 95, MCP 50). Four habitat types (grassland, open woodland, open forest, and mudflats) were delineated on aerial photographs and a habitat map produced using ARC/INFO. Compositional analysis was used to examine habitat preference in cats. Home range sizes of cats (n = 15, 598 fixes) in winter 1995, prior to fox removal, were similar to those reported in the only published study of cat spatial use in Australia, but larger than those recorded elsewhere. This may have reflected more dispersed food resources in Australia, although home range size was not correlated significantly with rabbit abundance. Male ranges (MCP 95 = 284 ha, n = 11) tended to be larger than females ( = 151ha, n = 4), but no differences were detected between young (1-3 years, = 271ha, n = 7) and old (>3 years, = 221ha, n = 8) cats. Cats were active both by day and night with no temporal differences being detected in range size. Both adult male and female cats tended to be solitary, although home ranges overlapped extensively. Kin groups were indicated (but not confirmed) as most inter-sexual overlap occurred between young and old cats. Habitat composition of home ranges generally reflected the availability of habitats at the study site, although cats significantly avoided mudflats. Home ranges comprised mostly open woodland and open forest habitats with smaller areas of grassland and mudflats. However, within individual home ranges, cats used grassland and open woodland habitats most often where rabbits were more abundant. Inter-individual (sex, age) or temporal (day/night) differences in habitat use were not detected. Comparison of resource use between cats and foxes indicated a large overlap in diet, home ranges and habitat use. Dietary breadths and overlaps between cats and foxes increased when rabbit availability declined in autumn and post-RCD. Dietary overlap was high overall (75%), although some resource partitioning was detected. Rabbits were more important in the diet of cats than foxes, particularly in summer, when foxes ate more grasshoppers. Carrion, invertebrates and vegetation were more important for foxes than for cats overall. Home ranges of both cats and foxes comprised mostly open woodland habitats followed by grassland, open forest and mudflats, which largely reflected their relative availabilities. However, within individual home ranges, cats showed a preference for grassland habitats. In addition, cats tended to deposit scats more often than foxes at rabbit warrens and at hollow log entrances, while foxes deposited scats more often than cats on sand plots, tracks and at dams. The large overlap in resource use between cats and foxes indicated a high potential for exploitation competition. Foxes may attempt to lessen competition by killing cats (interference competition). Three radiocollared cats were killed by foxes and aggression was observed toward cats. Home ranges overlapped extensively, but avoidance was indicated from the simultaneous radiotracking of both predators, as greater separations and lower overlaps in home ranges and core areas were recorded between species than within species. In addition, video observations suggested avoidance of carcasses by cats in the presence of foxes. The hypothesis that foxes limit feral cats through interspecific competition was then tested using a fox removal experiment. Foxes were reduced at two of the four sites from October 1995 using �1080� baiting and spotlight shooting by the VBCRC Predator-Prey project. Resource use and abundance of cats were compared before and after fox removal and between treated and untreated sites. Although no increase in cat abundance followed the removal of foxes, significant behavioural changes by cats strongly suggested interspecific competition operating via exploitation and interference. Exploitation competition was supported by the increased consumption of carrion by cats at the treated sites after fox removal, while support for interference competition came from the increased use of grassland habitats at night after fox removal. The direction of the resource shifts to more prey-rich habitats indicated asymmetry in the relationship between the two predator species. Although the null hypothesis of no limitation of cats by foxes could not be rejected, as no increase in cat abundance was recorded after fox removal, interspecific competition was considered to be the most likely mechanism limiting feral cats at Burrendong. Intraguild predation was not indicated as no cat remains were found in any of the 343 fox scats or 255 fox stomachs that were examined. In addition, minimal evidence was found for facilitation between cats and foxes, or for food limitation. The potential for foxes to limit cats, as shown in this study, indicates that cats need to be considered in future fox control operations. Integrated pest management, where foxes, cats and rabbits are controlled together, is strongly proposed if the objective is to safeguard native fauna in Australia. Further research is required to improve the effectiveness of current techniques for censusing cat populations, particularly in forested areas. This is essential for monitoring the effectiveness of control campaigns and quantifying factors that limit cat populations, and ultimately for effective protection of susceptible native fauna.

Researchers have found that inter-firm collaboration, that is, co-operative business networks, can provide a competitive advantage that would not be possible independently for small sized firms. Work has been done by some governments, for example, Danish, Norwegian, New Zealand, American and Japanese, in the area of policy and practice of business networks because they have realised the importance of business networking and have encouraged collaboration of small firms by assisting in the facilitation of networks. The Australian government established a Business Network Program which ran for four years and several studies were completed on various aspects of the program during that period. However, there had been no particular research that examined the success or other outcomes of these networks, thus providing the basis for the research question addressed in this research: How and why did the business networks developed in the AusIndustry Business Networks Program, succeed or not succeed? Further, questions relating to how and why these outcomes may have occurred or how they may have been measured in the Australian government facilitated program were also unanswered. A review of the extant literature in this area established the theoretical foundations upon which this research is based and made possible the development of a model comprising three constructs or research issues that would address the research question: RI 1: How and why is network success evaluated? RI 2: How and why do the internal and external environments affect the outcomes of the network? RI 3: How does facilitation affect the network? In order to address these research issues and the research question, a protocol was developed and case study interviews with the lead business of sixteen networks participating in the AusIndustry Business Network Program were carried out. The resultant data was compared for each of the research issues through a qualitative methodology from which conclusions and answers to the research question and issues were derived. The results of this research showed that network members evaluated their own outcomes often using multiple measures, both qualitative and quantitative, with the most common criteria being whether the network continued or discontinued. Moreover, it was concluded in this research that successful networks usually had a single goal or purpose for joining a network which they ultimately achieved. In contrast the unsuccessful networks generally joined the network with multiple goals and which were not all achieved, thus contributing to their lack of success. This result was not evident in the literature reviewed in chapter 2. Additionally, the findings showed that high levels of trust, commitment and reciprocity were essential elements in the success of business networks. More importantly this study found that whilst all successful networks had these elements, some of the non successful ones also reported high levels of trust, commitment and reciprocity. Thus it appeared in this study that whilst these elements are important for network success, they do not alone ensure that success, further, it was noted that for any network that reported a lack of any one of these elements, non success was more likely. In relation to this finding was the discovery that in these networks formal contracts between the network members increased the levels of commitment and reciprocity and thus increased the chances of success. When external environmental factors were examined in relation to their impact on network success, it was found that whilst all had some impact on their business generally, competition was noted as having the highest impact and government or legal issues the lowest impact. Finally, this research found that facilitation did not necessarily contribute to a network’s success but that possible a lack of appropriate facilitation style did contribute to the non-success of networks. However, it was clear that the small networks needed less facilitator guidance overall and that the larger networks definitely needed facilitation and guidance. Moreover, it was found that the type of facilitation at the various stages of the network process were more important to the likelihood of success, rather than the mere presence of a facilitator. Thus, the main contribution of this theory building research is to extend the general level of knowledge about business networks and provide new insights into network theory and the value of networks using an original application of existing knowledge. This knowledge can contribute to network education and training in business schools and can contribute to the development of future government policy and practice pertaining to network programs.

Thesis (Ph. D.)--Oregon State University, 1989. / Typescript (photocopy). Includes bibliographical references. Also available on the World Wide Web.

Thesis (Ph. D.)--Oregon State University, 1989. / Typescript (photocopy). Includes bibliographical references. Also available on the World Wide Web.

Thesis (Ph. D.)--University of North Carolina at Chapel Hill, 2007. / Title from title screen (viewed Aug. 13, 2008). Advisors: Patrick Conway ... [et al.]. "... in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Economics." Discipline: Economics; Department/School: Economics. Includes bibliographical references.

Dissertation (Ph.D.)--Auburn University, 2006. / Abstract. Vita. Includes bibliographic references.

Thesis (Ph. D.)--Oregon State University, 1990. / Typescript (photocopy). Includes bibliographical references. Also available on the World Wide Web.

Thesis (Ph. D.)--University of Adelaide, Dept. of Soil Science, 1996. / Copy of author's previously published work inserted. Includes bibliographical references (leaves 261-306).

Thesis (M.S.)--Oregon State University, 1989. / Typescript (photocopy). Includes bibliographical references. Also available on the World Wide Web.

Thesis (B.A. (Hons.))-- University of Adelaide, Dept. of Psychology, 1977.