Abstract Prospective and retrospective epidemiological studies have been conducted to study the relation between benign prostatic hyperplasia (B.P.H.) and prostate cancer. In the prospective study, 296 B.P.H. cases and 299 age-matched controls were traced until death or Dec. 31, 1972. The age-adjusted death-rate from prostate cancer was 3·7 times higher in the B.P.H. group than in the controls. For the retrospective study, hospital records of 290 prostate-cancer cases and 290 age-matched controls were compared. Matched-pair analysis of the antecedent hospital admissions revealed a relative risk of 5·1 for prostate cancer in patients seen earlier for benign prostatic enlargement. These findings suggest the need for a controlled clinical trial of prostatectomy as a preventive measure for prostatic cancer in symptom-free B.P.H. patients who have additional epidemiological characteristics associated with a high risk for prostate cancer.