1 Pharmacokinetics of pindolol were studied in normal subjects given 5, 10 and 20 mg orally and 3 mg i.v. Plasma half time was 2.9 +/- 0.3 (s.e. mean) h for both routes; peak drug levels occurred 1--2 h after ingestion and bioavailability was 53%. Plasma protein binding was 38% and was independent of plasma concentration; the drug was not concentrated in the red cell. 2 Work-heart rate regression lines were calculated from resting heart rate and three grades of 'steady-state' exercise standardized for the maximum work capacity (Wmax) of each subject. The equation was characterized by slope and HR50 (calculated heart rate at 0.5 Wmax). 3 After giving 5 mg i.v. pindolol to produce maximum cardiac beta-adrenoceptor blockade there were differences in inhibition of resting heart rate, slope, HR50 and maximum heart rate suggesting differences in sympathetic components. However, estimates of the degree of inhibition were closely similar for each variable when determined before and after atropinization indicating that the accuracy of estimation was independent of the level of vagal activity. 4 After oral pindolol peak inhibition of resting heart rate, slope and HR50 coincided with peak plasma concentration. Peak reduction of resting heart rate was greatest at the lowest dose, but inhibition of slope and HR50 were similar at all doses. 5 The different heart rate parameters recovered at different rates. After 24 h slope had returned to control, and the residual inhibition of HR50 reflected residual beta-adrenoceptor blockade of resting heart rate, as demonstrated by a shift in isoprenaline-heart rate relationship. 6 Inhibition of HR50 and other exercise parameters were 20% less in the concentration range 5--20 ng/ml than peak inhibition obtained in the range 21--160 ng/ml. The higher potency of pindolol compared with propranolol can be accounted for by the difference in protein binding.