A survey of current approaches to the development of vaccines against enteric diseases illuminates several strategies for the rational design of mycobacterial vaccines. One approach is to identify virulence factors and, through the use of transposon-mediated gene inactivation, to produce mutants that lack these factors. Such a strategy can provide important information on antigens important to protection. Genetic deletion or modification of these virulence factors provides insight into both the mechanisms of pathogenesis and the possibilities for the development of live attenuated vaccines. Alternatively, individual antigens identified as important for protection by antibodies and T cells or by studies of pathogenesis can be introduced into other live vaccine vectors to generate multivaccine vehicles. Both approaches appear to be relevant and of possible use for the development of improved vaccines against tuberculosis.