This study was designed to assess the relative antiarrhythmic activity of fixed doses of 2 beta-adrenergic blocking agents, propranolol and acebutolol, in a prospective double-blind crossover trial. Twenty-one patients who had at least 30 premature ventricular contractions (PVCs) per hour while receiving placebo were entered into the study. Ten patients were randomized to initially receive propranolol, 40 mg every 8 hours, and 11 were assigned to receive acebutolol, 300 mg every 8 hours. After 6 weeks of treatment, patients were weaned off medication for 1 week and then placed on placebo for 1 week. Eighteen patients were available and eligible for crossover to the alternative regimen for an additional 6 weeks. All 21 patients completed courses with propranolol and 17 completed courses with acebutolol. The mean number of PVCs per hour during placebo, propranolol and acebutolol treatment were 267, 87 and 119, respectively. Using paired t test statistics on observation differences, both propranolol and acebutolol significantly reduced the number of PVCs per hour compared with placebo, whereas similar analysis revealed no significant difference in the antiarrhythmic effect. However, with the current sample size the power of the test is too low for the latter conclusion to be stated with confidence. Side effects were mild and infrequent, requiring discontinuation of acebutolol in 2 patients and discontinuation of propranolol in 1. Thus, acebutolol is a safe and effective antiarrhythmic agent and compares favorably with propranolol.