A continuous ECG recording has been made in 31 myocardial infarction patients during the first 24 hours after admission to hospital. The number and severity of ventricular arrhythmias were recorded in great detail. Before discharge from hospital the patients were submitted to 20 hours of ECG tape recording, an exercise test on a bicycle ergometer and a static work test (handgrip). Another exercise test was performed one month after discharge. During the first day in the Coronary Care Unit (CCU) all 31 patients had ventricular arrhythmias and in 27 of them the arrhythmia was classified as major (calling for treatment according to Lown's criteria). At the exercise tests 23 patients showed ventricular arrhythmias, 12 of them considered as major. No antiarrhythmic therapy was given during the investigation. No correlation was found between the degree of arrhythmia during the first day in the CCU and during the exercise tests. Tape-recorded ECG's appeared to be inferior to dynamic exercise tests in the ability to disclose a latent tendency to ventricular arrhythmia. Static work did not provoke any ventricular arrhythmias. At a 2-year follow-up 5 patients had died, 4 of them suddenly. Examination of additional material on 11 patients with ventricular tachycardia or ventricular fibrillation during the CCU stay, showed that 2 had died, but only one suddenly. Frequency and severity of arrhythmias during the first day after the infarction seemed to correlate poorly to a persistent tendency to arrhythmias during the first day after the infarction seemed to correlate poorly to a persistent tendency to arrhythmias or to the risk of sudden death during the following 2 years. A dynamic exercise test performed before discharge would appear to be more effective in selecting patients in need of long-term prophylaxis. However, very few patients seem to need such a specific antiarrhythmic prophylaxis.