A group of 496 patients was entered into a randomized trial to study the possible value of fluorouracil in conjunction with surgical resection of carcinoma of the colon. Patients randomized to receive drug were given fluorouracil, 12 mg/kg body weight intravenously, for five successive days beginning 14 days after operation. Those termed clinically "curative" (A) and those termed clinically palliative (C) received a second course of drug six weeks later. Those termed proved palliative (B) continued to receive an intermittent course of fluorouracil. At 4½ years postrandomization, life table survival in group A, "currative resection," is 56% in treated patients and 50% in controls. Survival at 18 months in group B is 30% in treated vs 16% in controls, and group C, 57% in treated vs 31% in controls. While these results are encouraging, sufficient numbers of patients have not been followed up for a long enough period to justify valid evidence of drug benefit.