The effects of the cardioselective .beta.-blocker, metoprolol, were evaluated under double-blind conditions in 18 patients with angina pectoris. During an introductory run-in period of 8 wk, a placebo was given single-blindly. Thereafter 2 double-blind crossover periods each of 4 wk followed, either 20 mg metoprolol or placebo being given tid[3 times daily]. Metoprolol gave a significant reduction in the number of anginal attacks and in nitroglycerin consumption. The patients'' subjective assessments of their daily angina pectoris symptoms also showed a significant improvement compared with the placebo. At the end of each period, a standardized exercise test was performed. In comparison with placebo, metoprolol gave a significant increase of total work performed until the appearance of 1 mm ST-segment depression and until the end of exercise. The heart rate was significantly reduced at rest and during exercise. The blood pressure was significantly reduced only during exercise. None of the patients reported any severe unwanted effects. The complaints reported were mild to moderate, and the frequency during metoprolol treatment was even lower than during placebo treatment. No signs or symptoms of cardiac failure were seen in any of these patients on any occasion. Metoprolol ( 20 mg tid) is of benefit in the treatment of angina pectoris but further benefit might be obtained with higher doses.