Bovine tuberculosis is a global cause for concern in livestock, free-ranging wildlife, zoological collections and the human population. Large amount of time, effort and resources are spent on its diagnosis and control methods. This study was aimed at determining the sensitivity and specificity of the IS6110 specific PCR test on formalin fixed, paraffin embedded (FFPE) tissue blocks, compared to that of the gold standard method culture and to differentiate M. bovis from other members of the M. tuberculosis complex using the RD4 region of difference specific PCR test. A total of 141 FFPE tissue blocks of wild animals from game reserves, the National Zoological Gardens and routine tuberculosis (TB) surveys in Kruger National Park were tested. Among the 50 known TB positive samples (35 M. bovis culture positive, twelve M. tuberculosis culture positive and three diagnosed tuberculosis positive on histopathology examination) the IS6110 PCR had an overall sensitivity of 22%. The positive predictive value of the IS6110 test (91.67%) was quite high implying that although sensitivity was low, one can be highly confident that a positive test result is a true reflection of the positive disease status. The overall sensitivity of the RD4 PCR was 20%. The positive predictive value of the RD4 test (41.67%) was low, implying that a positive test result may be unreliable. The sensitivities of the M. tuberculosis and M. bovis culture positive samples were compared and a significant difference was noted. Sensitivities of the IS6110 and RD4 assays in M. tuberculosis culture positive samples were 66.67% and 33.33%, respectively; sensitivities of the IS6110 and RD4 assays in M. bovis culture positive samples were 8.57% and 17.14%, respectively. Difference in bacterial load in tissues infected with the two mycobacterial species may account for this finding (i.e. M. bovis infections have a lower bacteria load). Of the 91 known TB negative samples, the specificity of the IS6110 (98.90%) and RD4 (84.62%) PCR tests were high, but the negative predictive values of 69.67% and 65.81%, respectively, suggest that the probability of negative test results being incorrect still exists. The resultant sensitivity was increased when parallel interpretation was applied to histopathology examination and the IS6110 or RD4 PCR tests and when applied to the IS6110 and RD4 PCR tests. Both histopathology examination and PCR tests produce rapid results and their combination can be used in routine diagnostics. The RD4 PCR assay was unable to distinguish M. bovis from other members of the MTB complex and based on the findings of this study the RD4 PCR cannot add value to the diagnosis of suspect tuberculosis samples at this stage, but successful troubleshooting relating to 1) extraction method, 2) DNA inhibitors, 3) contamination and 4) multisampling protocol, may enable its use in future. / Dissertation (MSc)--University of Pretoria, 2013. / gm2014 / Veterinary Tropical Diseases / Unrestricted