The non-steroidal anti-inflammatory drug, sulindac, inhibits the growth of colorectal tumours in animal models of colon cancer and causes regression of polyps in patients with familial adenomatous polyposis. The mechanism by which sulindac exerts this inhibitory effect is not known, but it has been postulated to be via the inhibition of prostaglandin synthesis. However, two recent studies have indicated that sulindac sulphone, the non-prostaglandin inhibiting metabolite of sulindac, may be important in tumour inhibition. In the present study, we examined the effect of sulindac sulphone on the formation of aberrant crypt foci, the earliest identifiable lesions in the development of colorectal cancer, in the rat colon. We have previously shown that sulindac causes a dose dependent inhibition of aberrant crypt formation in this model. Aberrant crypt foci were induced with two oral doses of 1,2-dimethyl hydrazine at 25 mg/kg per dose. Treatment with sulindac sulphone at either 10 mg/kg b.d., or 20 mg/kg, b.d., was started on the day following administration of the first carcinogen dose and was continued for 3 weeks. Colons were then removed and examined for aberrant crypt foci. Colonic crypts were visualized by staining the unsectioned colon in 0.2% methylene blue solution. There was a significant reduction in the number of aberrant foci in rats treated with sulindac sulphone at 20 mg/kg, b.d. (ANOVA, P = 0.0054). The mechanism by which non-steroidal anti-inflammatory drugs inhibit formation of aberrant crypt foci is not clear; however, these data suggest that it is not due to the inhibition of prostaglandin synthesis.