Ten normal subjects participated in a placebo-controlled, randomized, parallel study to determine the effects on thyroid hormones of chronic (4 wk) propranolol or nadolol, including observation for 2 wk after their discontinuation. Subjects took placebo for 1 wk, then propranolol or nadolol doses increased weekly to 240 mg/day by 3 wk. After 1 wk of placebo, after 2 wk of the highest dose of propranolol or nadolol, and 2, 4, 6, 9, and 13 days after their discontinuation, thyroid hormone levels were measured by radioimmunoassay and heart rate responses to exercise were assessed. Both drugs induced equal and high degrees of exercise tachycardia inhibition. Propranolol decreased 3,3'5-triiodothyronine (T3) levels, increased 3-3'-5'-triiodothyronine (rT3) levels, tended to increase thyroxine levels, but did not increase thyroid-stimulating hormone levels. After discontinuation of propranolol, rT3 levels slowly (day 6) returned to values after placebo, suggesting delayed recovery of 5'-deiodination. There was no evidence of any rebound in T3 levels after withdrawal of propranolol. Nadolol induced no significant changes in the thyroid hormones measured. The data agree with the known effects of propranolol on thyroid hormones in normal man and show that nadolol does not have these effects when given chronically at an equivalent beta-blocking dose. The likely explanation is that the membrane-stabilizing activity of propranolol alters thyroid physiology by interfering with 5'-deiodinase.