We studied 76 patients with first recent myocardial infarction being not older than 12 hours. The patients included 58 men and 18 women. Their mean age was 62 years. We recorded continuously during the first three days following infarction the heart rate, all forms of ventricular premature beats, ventricular tachycardia, ventricular fibrillation, clinical status and activity of creatinine-phosphokinase and its isoenzyme MB. The results showed that ventricular premature beats (coupled and multiform) as well as ventricular tachycardia were more frequent in the first day of illness, while ventricular premature beats (except bigeminy, for which there is no explanation) were infrequent in the second and the third day after development of an infarct. The incidence of ventricular tachycardia during the follow-up period did not differ significantly. Ventricular fibrillation developed in 7 patients (9.2%). A comparison of the relation between ventricular premature beats and malignant ventricular tachycardia, i.e. ventricular tachycardia and ventricular fibrillation, revealed that the patients with more frequent ventricular tachycardia usually had frequent ventricular premature beats, particularly more often bigeminy, trigeminy, polymorphous ventricular premature and coupled ventricular premature beats, but not ventricular premature beats with R-on-T phenomenon. Ventricular tachycardia, however, was also found in patients with an evidence of more rare ventricular premature beats. This suggests that the occurrence of aforementioned forms of ventricular premature beats denotes only a somewhat greater probability that ventricular tachycardia will occur. The fact that there is a lack of correlation between ventricular tachycardia and R-on-T phenomenon indicates that this probability is not so significant. In conclusion, the authors believe that the patients with recent myocardial infarction and ventricular premature beats should be adequately followed up, and that prophylactic antiarrhythmic therapy is not required in most cases, as it was previously widely accepted concept. It should be administered only when ventricular tachycardia develops. Patients with ventricular fibrillation had more frequent ventricular premature beats, although ventricular premature beats in these patients were not statistically more frequent from those found in the patients in whom ventricular fibrillation was not verified. The presence or absence of ventricular tachycardia and ventricular fibrillation, respectively, had no influence on the other followed up parameters.