The past decade has seen an explosive increase in the development of potential new tuberculosis vaccine candidates, as well as the establishment of at least two testing centers. Various animal models, but particularly the mouse and guinea pig models, have provided a lot information about how new vaccines can reduce disease progression and how this influences the pathology of the disease, but there is still much to learn at the immunological level, particularly in terms of the nature of the T cell response that is needed to confer long lived resistance. Several categories of vaccine candidates have been tried to date, and there are at least five individual vaccines moving towards clinical evaluation. There are still areas of the field that are poorly developed however. These include the fact that we have no models of post- exposure vaccination, or any models of latent disease. In addition, no standardized models of safety/toxicology exist as yet, which will be needed before extensive clinical development of the new vaccines.