The effects of hydroxychloroquine and sulphasalazine on progression of joint damage shown by X-rays were compared in a double-blind, randomised trial in 60 patients with rheumatoid arthritis not previously treated with slow-acting antirheumatic drugs. X-rays of the hands and feet at the start and after 24 and 48 weeks of treatment were available for 28 patients treated with hydroxychloroquine and 22 treated with sulphasalazine. Erosions and joint space narrowing were scored by a single observer unaware of treatment. At baseline there were no significant differences in demographic, clinical, or radiographic characteristics between the treatment groups. Patients withdrawn because of lack of effect were included in the analysis. The median number of erosions was lower in the sulphasalazine than the hydroxychloroquine group at 24 weeks of treatment (2.5 vs 10) and the difference was significant at 48 weeks (5 vs 16; p less than 0.02). The difference in median total score of joint damage was significant at 24 weeks (6.5 vs 17; p less than 0.02) and at 48 weeks (8 vs 33; p less than 0.02). The increase in number of erosions and total score was significantly greater in the hydroxychloroquine than the sulphasalazine group, both after 24 weeks and after 48 weeks of treatment.