To investigate whether resistant starch (RS) affects putative risk factors for colon cancer, 24 healthy men consumed a daily RS supplement for 4 wk in addition to their habitual diet in a single-blind, randomized, balanced multiple crossover trial. During the first week, all subjects consumed the control supplement containing glucose. Subsequently, each subject consumed, in random order, a supplement with RS2 (uncooked high-amylose cornstarch), RS3 (extruded and retrograded high-amylose cornstarch), and glucose, each for 1 wk. The RS2 and RS3 supplements provided 32 g RS/d. Lithium was added to the supplements to measure compliance. Feces, 24-h urine, and breath samples, as well as a 24-h food-consumption recall were obtained weekly from each subject. Compliance as measured by urinary lithium recovery was satisfactory. The mean composition of the background diet did not differ between the various supplementation periods. Breath-hydrogen excretion, stool weight, and fecal starch excretion were significantly higher during RS than during glucose supplementation, but did not differ during RS2 and RS3 supplementation. There were no significant differences in fecal dry weight, pH, or short-chain fatty acid concentrations, nor in the pH, bile acid concentrations, cytotoxicity, or osmolality of fecal water. It is concluded that in healthy men, supplementing the habitual diet for 1 wk with 32 g RS2 or RS3/d compared with glucose had no effect on putative risk factors for colon cancer, except for increasing stool weight and colonic fermentative activity. There were no significant differences between the effects of RS2 and RS3 on the indexes studied.