The success of Mycobacterium tuberculosis as a pathogen is largely attributable to its ability to persist in host tissues, where drugs that are rapidly bactericidal in vitro require prolonged administration to achieve comparable effects. Latency is a frequent outcome of untreated or incompletely treated M. tuberculosis infection, creating a long-standing reservoir of future disease and contagion. Although the interactions between the bacterium and its host that result in chronic or latent infection are still largely undefined, recent years have seen a resurgence of interest and research activity in this area. Here we review some of the classic studies that have led to our current understanding of M. tuberculosis persistence, and discuss the varied approaches that are now being brought to bear on this important problem.