A group of 22 newly diagnosed noninsulin-dependent diabetic subjects and seven nondiabetic subjects underwent a glucose clamp at plasma glucose 100 mg/dL with insulin infusion rates of 1.0 and 10 mU/kg/min. During both insulin infusion rates, there was a sustained rise in plasma growth hormone (GH) above basal in 18 of the 22 diabetic subjects. Basal GH values were 2.37 +/- 0.67 ng/mL, rising above basal during the lower insulin infusion (6.1 +/- 3.3 ng/mL, P = 0.05) with a further rise at the higher insulin level (8.58 +/- 2.0 ng/mL, P less than 0.001). There was no rise in GH in any of the nondiabetic subjects. In neither group was there any rise above basal in cortisol, prolactin, glucagon, or somatostatin (SRIH). In a group of three nondiabetic subjects, a rise in GH similar to that seen in the diabetic group was induced by elevating the plasma glucose to 200 mg/dL for 60 minutes prior to the euglycemic clamp procedure. However, it is unlikely that changes in plasma glucose account totally for the changes in plasma GH described in the diabetic subjects since a rise in plasma GH was also seen in four diabetic subjects clamped at their fasting plasma glucose. We conclude that in newly diagnosed noninsulin-dependent diabetic subjects there is a rise in plasma GH during the euglycemic clamp procedure, which may be due to both the prior lowering of plasma glucose and the high plasma insulin levels.