The impact on plasma glucose of 35 g of fructose or an equicaloric amount (43.75 g) of high-fructose corn syrup (HFCS) (as part of a 400-calorie meal) was measured in six patients with non-insulin-dependent diabetes mellitus (NIDDM). Blood samples were collected periodically at all points for all patients for 3 h for plasma glucose (PG) and insulin (IRI) determinations. The mean peak PG increment was higher after the HFCS meal (66.5 mg/dl) than after the fructose meals (45.5 mg/dl). When increase in the mean plasma glucose concentration (delta PG) after the fructose meals were compared with the delta PG after the HFCS meals, there was statistical significance at 15 min (P less than 0.02) and 30 min (P less than 0.05). The total areas under the 3-h curves of mean delta PG showed a highly significant (P less than 0.001) difference between the fructose meal (5601 planimetry U) compared with the HFCS meal (8023 planimetry U). Mean changes in IRI after meals with either sweetener were comparable. These findings suggest that fructose is superior to HFCS as a sweetening agent in patients with NIDDM.