An increased supply of FFAs for oxidation leads to a reduced rate of glucose oxidation and interferes with the inhibitory action of insulin on hepatic glucose production. Available evidence indicates that in humans skeletal muscle is a site for such substrate competition, which involves both pyruvate oxidation and glycogen synthesis. The insulin resistance of obesity is thought to be mostly of metabolic origin, and fully reversible. A reduction in FFA supply by weight reduction can, however, reverse this defect. The insulin resistance associated with NIDDM is thought to be primary, with a strong genetic basis, and partially irreversible. Patients with NIDDM are unable to increase their glucose oxidation normally in response to insulin to meet the energy demands of the body. Increased oxidation of lipids represents a compensatory phenomenon to meet these demands. Therapeutic use of the glucose-FFA cycle to lower blood glucose levels has yielded conflicting results. Studies are in progress to develop agents that inhibit gluconeogenesis by interfering with FFA oxidation. Nicotinic acid derivatives seem to enhance glycogen synthesis acutely by activating glycogen synthetase. Whether these or similar agents can be used to restore impaired glycogen synthesis, the most characteristic genetic defect in NIDDM, cannot be answered until the effect has been proven in chronic studies. The existence of substrate competition between amino acids and glucose, and an intrinsic hypoaminoacidaemic property of amino acids, makes it possible to expand the Randel cycle into a glucose-FFA-amino acid cycle, which integrates control of substrate disposition at the whole body level.