Limiting factors of maximal exercise performance are not clearly defined. In order to differentiate between various factors, maximal exercise was studied during continuous (n = 12) and intermittent (n = 9) exercise. The non-selective beta-blocker timolol (10 mg b.i.d. for 5 days) was compared double-blind and placebo controlled with the beta-1-selective beta-blocker metoprolol (100 mg b.i.d. for 5 days), with respect to effect on maximal exercise tolerance. Total cumulated work was comparable during continuous and intermittent exercise. Timolol and metoprolol reduced maximal exercise performance. No difference was observed between the two beta-blockers during intermittent exercise. The non-selective beta-blocker caused a greater reduction in exercise performance (10.4%) than the beta-1-selective beta-blocker (4.7%) (P less than 0.05) during continuous exercise. Maximal heart rate was higher with metoprolol than timolol during continuous exercise. The non-selective beta-blocker caused a slightly greater inhibition of lipolysis than the beta-1 selective one. No significant differences in glucose concentrations were observed between the treatment regimens. Exercise caused a marked increase in serum potassium concentrations. Beta-blockade caused further increase in potassium at any given workload. This study indicates that maximal working capacity is comparable during continuous and intermittent exercise. Beta-1-selective and non-selective beta-blockade reduce the maximal working capacity, non-selective more than beta-1-selective. Substrate availability was not responsible for the beta-blocker induced reduction of the working capacity. The rate of rise in serum potassium was significantly higher during beta-blockade and may, therefore, be a limiting factor for the maximal working capacity.