The growing pandemic of human tuberculosis has not been affected significantly by the widespread use of the only currently available vaccine, bacille Calmette Guerin. Bacille Calmette Guerin protects uniformly against serious paediatric forms of tuberculosis and against adult pulmonary tuberculosis in some parts of the world, but there are clearly populations in high-burden countries which do not benefit from the current vaccination regimen. New tuberculosis vaccines will be essential for the ultimate control of this ancient disease. Research over the past 10 years has produced literally hundreds of new tuberculosis vaccine candidates representing all of the major vaccine design strategies; protein/peptide vaccines in adjuvants, DNA vaccines, naturally and rationally attenuated strains of mycobacteria, recombinant mycobacteria and other living vaccine vectors expressing genes coding for immunodominant mycobacterial antigens, and non-peptide vaccines. Many of these vaccines have been tested for immunogenicity and protective efficacy in mouse and guinea pig models of low-dose pulmonary tuberculosis. In addition, alternative routes of tuberculosis vaccine delivery (e.g. oral, respiratory, gene gun) and various combinations of priming or boosting an experimental vaccine with bacille Calmette Guerin have been examined in relevant animal models. One of the most promising of these vaccines is currently in Phase I trials in human subjects, and others are expected to follow in the near future. This review will summarise the most recent progress made toward the development and preclinical evaluation of novel vaccines for human tuberculosis.