Radiographic films of 40 patients participating in a single centre patient blind study of auranofin versus aurothioglucose were evaluated in a random order by one reader. The two treatment groups were comparable with respect to number of erosions and total radiographic score at the start of the study. Only in the auranofin-treated patients was a statistically significant increase in the mean number of new erosions (p less than 0.001 at 6 months and p less than 0.01 at 12 months treatment, paired t-test) as well as in the total radiographic score (p less than 0.01 at 6 and 12 months treatment, paired t-test) observed. Results of this study confirm that parenteral gold compounds do retard radiographic progression of joint destruction in the treatment of rheumatoid arthritis. The effects on radiographic progression shown in this study are in agreement with other reports which, based on clinical and biochemical parameters, have shown that auranofin is somewhat less effective than the injectible gold salts.