Two beta-blocking agents, non-selective propranolol and beta 1-selective metoprolol, were investigated with respect to their effect on glucose metabolism in 11 hypertensive, non-diabetic patients. They were randomly treated for two weeks in a double-blind cross-over manner with propranolol, metoprolol and placebo. Both drugs caused a small but significant increase in basal blood glucose values as compared with placebo (p less than 0.01). Metoprolol increased the blood glucose concentrations during the first 10 min of an i.v. glucose tolerance test (IVGTT) as compared with placebo (p less than 0.02) and propranolol (p less than 0.05). Propranolol raised only the blood glucose values during the later part of the IVGTT (p less than 0.01). The increase in blood glucose concentrations was, however, not associated with significant changes in peripheral insulin levels. The mean basal glucagon concentrations were lower during propranolol and metoprolol than during placebo (p less than 0.01). Propranolol also induced a more pronounced reduction of plasma glucagon than placebo (p less than 0.05) at 10 min of the IVGTT. The mean basal free fatty acid (FFA) concentrations were lower during propranolol (p less than 0.001) and metoprolol (p less than 0.05) than during placebo. Both drugs decreased the plasma levels of FFA during the first 10 min of the IVGTT as compared with placebo (p less than 0.01 and p less than 0.02, respectively). Pharmacological doses of propranolol and metoprolol increased blood glucose concentrations, decreased plasma glucagon and FFA concentrations, but had no effect on serum insulin levels in hypertensive, non-diabetic subjects.