Twenty-four adult men and women, classified as carbohydrate-sensitive on the basis of an exaggerated insulin response to a sucrose load, consumed diets containing 5, 18, and 33% of calories as sucrose for 6 wk each in a cross-over design. The diets contained identical natural and processed foods except for a patty containing 2, 15, or 30% of the calories as sucrose at the expense of wheat starch. Carbohydrate, fat, and protein provided 44, 42, and 14% of the calories, respectively. Of total calories, 25% were consumed at breakfast and 75% at dinner. Initial body weights of the subjects were essentially maintained. Fasting serum insulin levels increased with the sucrose content of the diet and were significantly higher in men than in women. Mean fasting glucose was significantly higher on either 18 or 33% sucrose than on 5% sucrose. The sucrose content of the diet did not affect fasting serum glucagon. When compared to the insulin response to a sucrose load (2 g/kg body weight) after consuming the 5% sucrose diet, serum insulin was significantly higher at 1 h after the 18% sucrose diet and at 0.5, 1, 2, and 3 h after the 33% sucrose diet. Except after 2 h, the glucose response was significantly greater after the 18 and 33% sucrose diets than after the 5% sucrose diet. These results indicate that sucrose intake by carbohydrate-sensitive individuals, even at levels approximating the average United States intake, can produce undesirable changes in several parameters associated with glucose tolerance.