The results of a controlled trial in 22 patients indicated that levamisole, administered in a dose of 150 mg per day for 2 months, was more effective than a placebo on the clinical symptoms and signs of rheumatoid arthritis. Side effects were seen in 8 of the 12 patients treated with levamisole. They necessitated the premature interruption of treatment in three, but were never serious (no cases of agranulocytosis were seen). In the group treated with levamisole, there was a slight decrease in sedimentation rate, a significant fall in levels of the C3 fraction of serum complement and of circulating immune complexes (detected by polyethylene glycol), and a significant increase in cutaneous reaction to tuberculin and to candidin. By contrast, there were no significant variations in levels of rheumatoid factor, serum immunoglobulins, C4 fraction of complement, lymphocyte transformation indices in the presence of six dilutions of phytohaemagglutinin, nor in the percentages of T and B lymphocytes. The effectiveness of levamisole in rheumatoid arthritis would appear to be demonstrated, but its mode of action remains open to discussion, in the absence of any definite correlation between therapeutic activity and the immunostimulant effects of the drug.