Starch that is resistant to human amylases forms during the cooking and subsequent cooling of some foods, and may therefore be a substrate for the bacterial flora of the colon. It is thus possible that resistant starch (RS) will affect colon function in a similar manner to non-starch polysaccharides. To test this theory, a group of eight volunteers took two diet supplements for 1 week each in a random order with a 1 week separation. One supplement comprised mainly 350 g Cornflakes/d and the other 380 g Rice Krispies/d, providing 10.33 and 0.86 g RS/d respectively. The amounts of amylase-digestible starch, nonstarch polysaccharides, total carbohydrate, energy, protein and fat were balanced between the two periods by giving small amounts of Casilan, wheat bran, butter and boiled sweets. The volunteers made faecal collections during day 3 to day 7 of each period. Whole-gut transit time was calculated using the continuous method. Stool consistency and ease of defaecation were assessed by the volunteers. All episodes of flatulence noticed were recorded in a diary, along with food intake. Serial breath hydrogen measurements were made at 15 min intervals for 8 h on day 1 of each supplement. Questionnaires regarding colon function were completed at the end of each dietary period. There were no significant differences in the stool mass, frequency or consistency, ease of defaecations, transit time or flatulence experienced during the two supplements (P greater than 0.05). Significantly more H2 (area under curve) was produced while eating Cornflakes than Rice Krispies (P less than 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)