A total of 123 out of 549 elderly residents of local authority welfare homes in Nottinghamshire were found at screening to have a standing or lying diastolic blood pressure of 100 mm Hg or more. These 123 subjects were randomly allocated to simple observation or to treatment with methyldopa. The cumulative mortality was similar in the observed and treated groups and in the normotensive group from which the subjects had been separated. Thus moderate hypertension, whether treated or not, was not a major risk predictor in the elderly population studied.