This trial comprised 40 patients with angina pectoris and reproducible ischaemic ST segment depression in the exercise ECG. Whenever possible the diagnosis of coronary heart disease was confirmed by coronary angiography. After a preliminary 7-day placebo period, 20 of these patients were treated for 4 weeks with 5-(3-tert-butylamino-2-hydroxy-propoxy)-3, 4-dihydro-2(1H)-quinolinone hydrochloride (carteolol hydrochloride, Endak, Endak mite), a nonselective beta-receptor blocker, having adrenergic properties (ISA) up to 30 times more powerful than those of propranolol. The trial was double-blind and randomized; standard medication (pindolol) was given to the controls. In general, the results obtained with carteolol--and also with pindolol--may be regarded as "good" or "very good". The average weekly incidence of anginal attacks fell from 6 to 4 (p less than 0.05). Carteolol produced a definite decrease in ST segment depression during exercise, at the end of exercise and in the recovery phase (p less than 0.05). The same was true for the patients treated with pindolol. There was no statistically significant difference between the treatment groups. Pulse rate and systolic blood pressure--measured before exercise and at the end of the recovery phase--did not change; however, readings made at the end of exercise showed a significant decrease (p less than 0.05). Diastolic pressure remained essentially unaltered. There were no changes in the chest radiograph or resting ECG. Laboratory results remained unchanged in both groups. In one patient of each group treatment needed to be discontinued because of dyspnoea, mainly during exercise. This reaction could probably be related to the beta-blocker therapy. The overall response to treatment, as assessed by the physician conducting the trial, was "good" or "very good" in 75% of the patients receiving carteolol and in 70% of those receiving pindolol. 70% of the patients who had received carteolol assessed the therapeutic result as "good" or "very good"; in the pindolol group 55% made the same judgment.