A tentative diagnosis of bovine tuberculosis can be made following the macroscopic detection at necropsy of typical lesions. Histo-pathological examination of the lesion may increase the confidence of the diagnosis but bacteriological isolation of Mycobacterium bovis from the lesion is the only way to make a definitive diagnosis. The sensitivity of gross post mortem examination is affected by the method employed and the anatomical sites examined. Careful examination of as few as 6 pairs of lymph nodes, the lungs and the mesenteric lymph nodes can result in 95% of cattle with macroscopic lesions being identified. Although during post mortem inspection of carcases at abattoir all the principle sites where lesions are to be found were examined, this procedure was found to be insensitive for the detection of lesions. To determine the significance of cattle that give a positive reaction in diagnostic tests but do not have visible lesions (NVL), a bacteriological examination is necessary. NVLs may be due to early infection, poor necropsy technique or infection with mycobacteria other than M. bovis. M. bovis was found to survive best in frozen tissue and the tissue preservative, sodium tetraborate, was found to have adverse effects on viability. It was found desirable to use two different culture media for the primary isolation of M. bovis; agar media for rapid growth and egg media for control of contamination. Additional control of contamination was achieved without adversely affecting the viability by treating the specimen before culture with 0.075% hexadecylpyridinium chloride. The addition of CO2 to the incubation atmosphere did not enhance the recovery of M. bovis. Conventional identification of isolates of M. bovis is by biochemical tests and cultural characteristics, but methods employing monoclonal antibodies and DNA probes may be used to obtain a rapid identification.