The increased understanding of the mechanisms which underlie rheumatoid disease has been accompanied by a more appropriate use of the limited repertoire of therapeutic agents. Conventional second-line drugs still have a role in everyday practice. The efficacy of these agents in reducing the severity of clinical signs of joint inflammation, whilst at the same time causing significant reductions in the laboratory measures of the acute phase response is undoubtedly confirmed by meta-analysis of several therapeutic trials of these agents. Whether or not these agents can influence outcome, usually assessed in terms of radiological progression, is more contentious. Furthermore, their toxicity in long term use is not inconsiderable. However, newer agents may play a more important part in therapy in the future. Such therapy can be designed to specifically interfere with the abnormalities of the immune system which characterise rheumatoid arthritis. Many of the agents reviewed have been successfully applied to animal models of arthritis but we still await large randomised controlled studies in humans to determine their clinical efficacy and toxicity. In view of the complexity of the immunological abnormalities in rheumatoid arthritis, it may be necessary to consider using a number of such agents in any particular patient. This should result in more rational therapy in rheumatoid arthritis.