Forty out-patients with acute 'flare' of chronic rheumatoid arthritis were treated orally with either 150 mg proglumetacin or 250 mg naproxen twice daily over 3 weeks, according to a randomized, double-blind design. Before and after 1 and 3 weeks of treatment, the number of painful and of swollen joints, the intensity of pain and function tests (morning stiffness, time to walk over 15 metres and hand grip strength) were measured and recorded. Haematology was investigated before and after treatment. Two patients in the proglumetacin group did not report to control and were considered drop-outs; 2 more (1 in each group) interrupted treatment before completion because of the onset or aggravation of accessory symptoms. Efficacy, assessed in 17 patients on proglumetacin and in 19 on naproxen, was good with both drugs, even though only those patients given proglumetacin experienced a significant (p less than 0.01) decrease in the number of painful joints. None of the haematological tests showed clinically significant variations after either treatment. Tolerance could be assessed in 18 and 20 patients given proglumetacin or naproxen, respectively. Accessory symptoms appeared or were aggravated in 5 and 3 patients, respectively.