In terms of the paradigms for antibacterial action presented in the introduction, there is good evidence that broad spectrum agents exert their anti-mycobacterial activity by interaction with classical targets occurring in a wide range of organisms including the mycobacteria. This is supported either by direct evidence (e.g., inhibition by rifampicin of mycobacterial RNA polymerase), or indirectly by the characterization of drug-resistant mycobacteria where mutations conferring resistance have been mapped to target sites homologous to those found in other bacteria (fluoroquinolones, macrolides, rifampicin, streptomycin). On the other hand, although the mode of action of some of the agents with an anti-mycobacterial spectrum is not fully understood, it is evident that the restricted spectrum is likely to arise from the possession of unique targets, or specific pro-drug conversion systems, or to a combination of both mechanisms. In several cases the narrow spectrum of the agents can be attributed to inhibition of molecular targets involved in the biosynthesis of the mycobacterial cell envelope that contains many unique polymers. The recent re-emergence of tuberculosis as an important human pathogen has led to improved methods for exploring the structure, biochemistry and genetics of the mycobacteria. These technical advances can now be used to gain a better understanding of the molecular basis of drug action in mycobacteria.