We have previously shown that fructose and sorbitol given with a standard meal cause less increment in plasma glucose than sucrose and high fructose corn syrup (HFCS) in patients with NIDDM. However, there was no direct comparison of sucrose with HFCS. Sixteen men and one woman aged 54–67) with NIDDM were given either 35 g sucrose, 35 g fructose, or 43.75 g HFCS containing 35 g carbohydrate as part of a 400-calorie test meal. Blood samples were obtained at frequent intervals up to 3 h and were analyzed for glucose and insulin. As compared with a fructose meal, the mean increment in plasma glucose (ΔPG) after a sucrose meal was significantly higher at 45 min and after an HFCS meal it was significantly higher at 30 and 45 min, but sucrose and HFCS meals did not differ. When delta PGs were compared in nine patients with basal PG > 140 mg/dl and in eight patients with basal PG < 140 mg/dl, differences in ΔPG after sucrose and HFCS versus fructose meals became more significant but still did not differ from each other. The integrated total areas under the ΔPG curves after sucrose, HFCS, and fructose meals were not statistically different. However, the areas under the curves up to 90 min after sucrose and HFCS meals, which did not differ, were greater than the fructose meal. The mean delta IRI after sucrose meals was markedly elevated at 45, 60, and 75 min (P < 0.05) and after HFCS meals at 45 min as compared with fructose meals. There was no significant difference in the mean AIRI between sucrose and HFCS meals. This study shows that sucrose and HFCS cause greater increments of PG than fructose in patients with NIDDM, but do not differ from each other. Thus, even though HFCS is less expensive than fructose, its effect on plasma glucose and insulin is not different from that of sucrose, and we cannot confirm on a scientific basis a useful function for HFCS in diets for persons with diabetes. Also, diabetic patients with higher basal PG (i.e., > 140 mg/dl) show similar increase in PG whether they are given sucrose, HFCS, or fructose meals. Therefore, fructose has potential value only in patients with mild diabetes mellitus.