Ambulatory blood pressure monitoring gives a more representative blood pressure profile than office blood pressure measurements and is free of any placebo effect. It is therefore useful for studying the effect of antihypertensive agents. Although ambulatory blood pressure is less variable than office blood pressure, spontaneous fluctuations have been found in whole-day blood pressure when repeated measurements are taken. In the multicentre Triveneto Study, the mean difference between 24-h blood pressure recordings taken 3 months apart in 85 mild hypertensives was -0.1/-0.7 mmHg and the coefficient of repeatability (2 s.d.) was 17.3/12.6 mmHg. The corresponding values for office blood pressure were -8.7/-2.0 and 29.8/16.5 mmHg, respectively. This reduction in inter-measurement variability with ambulatory blood pressure monitoring makes it possible to reduce the sample size required to prove the effect of an antihypertensive agent in pharmacological trials. However, in the individual subject, the results of ambulatory blood pressure monitoring should be considered with caution, as 24-h blood pressure averages and profiles are subject to a degree of variability. This technique was used in 21 mild-to-moderate hypertensives to test the antihypertensive effect of lacidipine given once a day (4-6 mg) versus placebo. The drug proved effective throughout the day and night, showing a 24-h effect on blood pressure without reflex tachycardia or other intolerable side effects.