Twenty-three out-patients with mild or moderate essential hypertension have been treated with a combination of hydralazine (37.5-150 mg daily) and oxprenolol (60 mg daily). Before treatment the patients were phenotyped for polymorphic acetylation by means of the sulphamethazine test: 12 proved to be slow and 11 rapid acetylators. A significant correlation was found between daily doses of hydralazine and the plasma hydralazine levels, separately in slow (r=0.480) and in rapid (r=0.580) acetylators. The antihypertensive response to hydralazine correlated well to plasma hydralazine levels. The mean fall of BP in slow acetylators was 33/23 mmHg in supine and 20/18 mmHg in standing position. The corresponding values in rapid acetylators were 22/15 and 21/15 mmHg. The average daily doses of hydralazine needed for these responses were 1.3 mg/kg in slow and 1.6 mg/kg in rapid acetylators. To reduce the systolic BP by 20 mmHg, 1.0 mg/kg of hydralazine was needed in slow acetylators; rapid acetylators needed a significantly higher dose of 1.4 mg/kg. During a follow-upof 1 year there have been virtually no side-effects. The results tally with the previous finding of Zacest and Koch-Weser, who demonstrated a similar correlation during the triple-drug regimen. It seems as if hypertensive patients can be succesfully treated with hydralazine and beta-blocking drug without knowledge of the patient's acetylator phenotype. However, acetylator status is a determinant of tissue levels and long-term toxicity of hydralazine, and patients should be phenotyped because beta-blockers may mask the warning side-effects.