To determine the contribution of skeletal muscle to fasting hyperglycemia in noninsulin dependent type II diabetes (NIDDM), the forearm balance of glucose, lactate, and alanine was quantified in 25 control subjects, 21 hyperglycemic (blood glucose: 11.6 mmol/L), and 19 insulin-treated patients with NIDDM (blood glucose: 5.8 mmol/L). Forearm glucose uptake was similar in controls (4.6 +/- 0.6 mumol L-1 min-1) and in hyperglycemic diabetic patients (4.5 +/- 0.9 mumol L-1 min-1). In spite of this, in the diabetic patients lactate (5.1 +/- 0.8 mumol L-1 min-1) and alanine (2.6 +/- 0.4) release by the forearm was 3- and 2-fold higher than in the control group (lactate: 1.7 +/- 0.8, P less than 0.005; and alanine: 1.3 +/- 0.2, P less than 0.05, respectively). The ratio of lactate release to glucose uptake was 57% and 18% in diabetic and control subjects, respectively. Insulin administration did not affect either glucose uptake or the release of gluconeogenic substrates by the forearm. It is concluded that: 1) in fasting patients with NIDDM, glucose is taken up by the skeletal muscle in normal amounts but preferentially used nonoxidatively with lactate formation. This suggests that, although the muscle does not contribute directly to fasting hyperglycemia, it may play an indirect role through an increased delivery of glucose precursors; and 2) insulin-induced normoglycemia is maintained by mechanisms that do not involve the exchange of glucose and gluconeogenic substrates by the skeletal muscle.