In this review the rationale for the possible beneficial effect of tetracycline derivatives for the treatment of rheumatoid arthritis is discussed. Early studies and the two open trials of the 1980s are briefly discussed. The three double-blind studies conducted in the 1990s (Kloppenburg et al., The Netherlands; the MIRA trial, USA and the O'Dell et al., USA) are described in detail. The baseline clinical and demographic data for these patients, as well as the efficacy and toxicity data are described in the text and summarised in tables. The long-term data of the O'Dell et al.'s trial is described. Finally, side effects not observed during the conduct of these trials, but reported to occur in other patients, for example those receiving minocycline for the treatment of acne, are also listed.