Rabies, one of the oldest diseases known to man, remains uncontrolled in many parts of the world. This is in spite of the availability of safe, effective and economical tools for its control. The wide-spread distribution and rapid growth of its primary reservoir host population in the developing world, the domestic dog, together with their ubiquitous association with human populations, explains why canine rabies is endemic across much of the globe. However, given that effective dog vaccines are available, and that rabies has been eliminated from dog populations in many countries (including large, non-island countries like the United States) through the implementation of these vaccines and other control measures, why does the disease continue to exact such a toll in other countries? One reason must be a lack of political and institutional will to tackle the problem. This in turn is based in part on a lack of awareness of the extent of the problem, coupled with competing interests for scarce public health resources. Reliance on the reporting of rabies cases via official channels may lead to underestimation of the true incidence of the disease by up to one hundred-fold, and a subsequent lack of prioritisation of resources for its control. This leads to a vicious circle of neglect - low priority means no resources available for surveillance, which in turn ensures that the true extent of disease occurrence remains unknown. It is hoped that the quantitative estimates of rabies burden presented in Chapter 2 will redress this, and provide impetus for policy makers and donors to tackle the problem. One advantage of the quantitative risk assessment method used in Chapter 2 is its transparency - as more detailed data become available so the model can be updated and refined. Knowledge of the estimated burden of the disease, together with preliminary data which shows that, in terms of cost per DALY saved, rabies is among one of the most cost-effective diseases to target (through the mass vaccination of dogs), have already encouraged international health agencies and donors to reconsider their prioritisation of this disease. In this respect, further research is needed to develop more refined models of cost-effectiveness estimates across different time horizons and under different epidemiological scenarios. An additional reason for the lack of rabies control in much of the developing world is that, while safe and cheap dog rabies vaccines are available, their effective implementation is often hampered by an incomplete understanding of the demographics and ecology of dog populations in rabies-endemic areas, and the relative roles of anthropogenic and non-anthropogenic regulatory factors. Accurate estimates of dog population numbers are rarely available, and existing estimates often grossly underestimate actual numbers of dogs. Chapter 3 provides a means whereby such numbers can be extrapolated for a variety of socio-cultural scenarios in Tanzania (including national estimates based on composite dog-human ratios), while at the same time highlighting the difficulties encountered in the estimation process (particularly the lack of sampling frames for households in the estimation of the owned dog population). National estimates of the owned dog population, derived from the work presented in Chapter 3, will be included in the national rabies control policy document currently being developed by the Tanzanian government, and have also been included in funding applications to donors to support this programme. These figures can be used to estimate required resources for rabies control efforts, and to allow the optimal utilisation of those 181 resources. They can also serve as targets against which vaccination efforts can be measured, to assess the vaccination coverage achieved. Humane dog population management programmes can serve as a valuable adjunct to vaccination efforts in the control of rabies, by decreasing dog population growth rates and reducing population turnover. Such stable populations greatly reduce the frequency of mass vaccination campaigns needed to maintain a given level of vaccination coverage, in addition to reducing (over time) the total number of dogs to be vaccinated. Chapter 5 provides valuable baseline data for an humane dog population control programme aimed at the sterilization of owned dogs in Colombo, Sri Lanka, and together with Chapter 3 provides a basis for comparison of owned urban dog populations in Asian and African environments. The characteristics of dog-owning households identified in these two chapters will enable the effective targeting of education and awareness campaigns, for both humane dog population programmes and rabies vaccination campaigns. The strong male bias identified in this study in the sex ratio of owned dogs, together with the preference for male dogs and the fact that few owners of female dogs would want to keep a litter, point to the existence of possible anthropogenic drivers of the population dynamics of both the owned and unowned dog population in Colombo. Further work is needed to understand the effect of these drivers, and their interaction with non-anthropogenic regulators. Such studies would need to examine the fecundity and sex-specific pup survival in both the owned and unowned segments of the dog population, and migration between the two sub-populations, through adoption or abandonment. This work is necessary to understand the effect of humane dog population 182 control programmes on dog population dynamics, and could be done in conjunction with studies on the impact of these programmes on the long-term cost-effectiveness of dog vaccination campaigns. Human behaviours which impact on the dynamics and welfare of dog populations are expressed as a result of the underlying attitudes of individuals towards dogs. These attitudes are in turn modified by socio-cultural conditions and individual experience. Chapters 4 and 6 present unique research into these attitudes and modifiers in a crosssection of inhabitants in Tanzania and Colombo. Building on the work done in Tanzania, the item scale developed in Chapter 5 provides a valuable tool with which these attitudes, and their relationship to dog welfare and rabies control, can be further explored. For example, the finding that household heads with more positive attitudes towards dogs were more likely to have their dogs vaccinated at a central point suggests the need for integration of programmes on responsible pet ownership into rabies vaccination planning. Overall, the work suggests a more positive affective attitude towards dogs in these countries than was perhaps previously supposed, and that this particular aspect of the human-animal interface may be worthy of further study. The effect of these attitudes, and the possible changes in attitude occurring as a result of the implementation of responsible pet ownership programmes, on the uptake and impact of the humane dog population control programme in Colombo, is a potentially fruitful area for further research. The application of the findings presented in this thesis to the refinement of vaccination campaigns for the control of canine rabies will be of benefit not only to the fields of public health and animal welfare, but also to the conservation of endangered canids threatened with spill-over of the disease from sympatric dog populations. The management strategies discussed in Chapter 7 for the reactive vaccination of Ethiopian wolf populations following such a spill-over event, although effective, are costly and potentially high-risk if implemented too late. Targeted vaccination of the dog population, ideally in conjunction with responsible pet ownership and humane population control programmes, may be more effective, and have the additional benefit of improving relations with local communities by reducing the incidence of human bites from suspect rabid dogs. Once again attitudes of owners towards dogs play a role, as the uptake and cost-effectiveness of these vaccination programmes is dependent on owner co-operation and influenced by the ability and willingness of owners to handle their dogs. Understanding these attitudes and their effects on owner behaviour may improve delivery of vaccines, possibly through the deployment of new technologies such as bait formulations containing oral rabies vaccines.