Blood-pressure screening in a family health centre identified 114 patients (53 male, 61 female) with diastolic pressures of 95 mm Hg and greater after three readings in the seated position (mean 163 [SEM 2]/104 mm Hg). All were instructed in the use of the 'Copal UA 231/251' electronic sphygmomanometer and produced a series of readings taken at home over 3 days. They were recalled after 2 weeks and 4 weeks for repeat clinic measurements of blood pressure. Blood pressure fell on successive clinic visits; at the final visit only 59 patients (31 male, 28 female) had diastolic pressures of 95 mm Hg or greater. Average day-time home blood-pressure measurements (155/94 mm Hg) were significantly lower than the screening blood-pressure measurements but were not significantly different from those at the third clinic visit (154/97 mm Hg). Home blood-pressure measurements were successful in predicting outcome at the third clinic visit in 90 (79%) patients; home-monitored pressures suggested normotension when the final clinic visit diastolic blood pressure was still above 95 mm Hg in only 16 (14%) patients. Only 2 of these had a final clinic diastolic pressure above 105 mm Hg. Home monitoring represents a practicable and acceptable alternative to repeated clinic measurements in the initial assessment of hypertensive patients.