Vaccination against bovine tuberculosis is likely to become an important disease control strategy in developing countries, which cannot afford a test and slaughter control programme, or in countries which have a wildlife reservoir of Mycobacterium bovis infection. In the past decade, considerable progress has been made in the development and evaluation of tuberculosis vaccines for cattle and for a range of wildlife maintenance hosts including possums, badgers, deer and African buffaloes. Experimental challenge systems have been established for the different target species and the resulting disease process has mimicked that seen in the field. In cattle, neonatal vaccination with BCG appeared to be more effective than vaccination of 6-month-old calves and in most situations no other vaccine has been shown to be better than BCG. However, prime-boost strategies involving combinations of BCG with a protein or DNA vaccine, to improve on BCG vaccination alone, have produced very encouraging results. Differential diagnostic tests have been developed using mycobacterial antigens that are only present in virulent M. bovis to differentiate between BCG-vaccinated and M. bovis-infected cattle. BCG vaccine has been shown to reduce the spread of tuberculous lesions in a range of wildlife species and a prototype oral bait delivery system has been developed. Prospects for the development of improved vaccines against bovine tuberculosis are promising and vaccination approaches could become very valuable in the control and eradication of bovine tuberculosis.