To determine whether epinephrine-induced hypokalemia is due to beta2-adrenoceptor stimulation, and whether hypokalemia can occur at physiologic concentrations of the agonist, epinephrine was infused into six normal volunteers at a rate of 0.1 microgram per kilogram of body weight per minute. The circulating epinephrine concentration was increased to 1.74 +/- 0.65 ng per milliliter, plasma potassium was reduced by 0.82 +/- 0.19 meq per liter, plasma insulin fell by 12 +/- 4 mU per liter, plasma renin activity was elevated, and tachycardia occurred. Isoproterenol infused at 0.02 micrograms per kilogram per minute caused similar tachycardia (25 beats per minute) and elevation in plasma renin activity (6.0 to 6.5 ng per milliliter per hour), but no hypokalemia. The difference in responses to the two catecholamines was ascribed to the relative beta2-selectivity of epinephrine. This hypothesis was tested in six subjects given infusions of epinephrine (0.05 micrograms per kilogram per minute) after administration of either 2.5 or 5 mg of ICI 118551--a selective beta2-receptor antagonist--or placebo. After placebo, epinephrine infusion elevated the circulating epinephrine concentration and reduced plasma potassium; hypokalemia was prevented by the beta2-antagonist. This drug only partially inhibited the rises in plasma renin and glucose and the shortening of systolic time intervals; there was no tachycardia. Fifteen-fold to 30-fold increases in circulating epinephrine concentration appear to cause hypokalemia by a specific beta2-receptor effect distinct from other actions of epinephrine. This phenomenon may be of physiologic importance after severe myocardial infarction, when similar increases in plasma epinephrine have occurred.