Rabies virus (RABV) primarily infects carnivorous terrestrial mammals in Africa. However, in 1977 the first epidemic of rabies in kudu occurred in Namibia. Due to the excessive spread and deaths of kudu, it was suggested that RABV was being maintained within the kudu population in a separate RABV cycle. Previous reports had suggested a massive increase in kudu population numbers – as well as densities - due to the game farming industry in Namibia, leading to a sufficient population threshold to be able to maintain a separate RABV cycle. In 1983 the epizootic ended, with an estimated loss of 30,000 kudu. A second epizootic began in 2002 and is still ongoing. The aims of this study were to lend support to the hypothesis that a separate RABV cycle is being maintained within the Namibian kudu population and that the RABV is adapting to its new herbivorous host. In addition, it was hypothesised that bovine viral diarrhoea virus (BVDV) may play a role in the susceptibility of Namibian kudu to RABV infection, due to the immune-suppressive effects of BVDV infection. Thus far, no other studies have determined phylogenetically, over the same spatial and temporal range, whether there is evidence for the separate maintenance of RABV in kudu. Furthermore, detailed molecular analyses were performed in order to determine whether RABVs isolated from kudu were diverging from isolates from canids and whether the mutations were under positive selection, showing selection pressures of the host on the virus. Results showed a clear phylogenetic differentiation of the isolates from canids to those from kudu, with all of the kudu isolates from a variety of geographical ranges clustering in a well-supported cluster, separate from canids from the same geographical ranges. Furthermore, full genome analyses showed several mutations unique (globally) to isolates from kudu, and other mutations separating isolates from kudu to those from jackals in Namibia, as well as several amino acids being under positive selection. Serological analyses of BVDV and RABV suggested no clear correlation between RABV and BVDV exposures, despite a high prevalence of BVDV in Namibia. In conclusion, this study showed strong supporting evidence that a separate RABV cycle is being maintained within the Namibian kudu population independently from any canid cycle. Furthermore, we showed no clear correlation of the immune-suppressive effects of BVDV on the susceptibility of Namibian kudu to RABV infection, despite a high prevalence of BVDV in Namibia. Lastly, it is suggested that further experimental trials are performed in order to determine the means of transmission of RABV among kudu, as well as further epidemiological surveys in order to determine the extent of the virus infection as well as risks that this RABV cycle poses to other herbivorous animals in Namibia and neighbouring countries.