The mean of rapidly repeated duplicate or triplicate measurements is often used in studies of antihypertensive drugs. Forty patients with hypertension had triplicate measurements of blood pressure and heart rate on two occasions, 1 week apart, during placebo treatment. The average difference between the first measurement and the mean of the triplicate measurements was -0.3 mm Hg. The average coefficient of variation for supine and standing, systolic and diastolic blood pressures was 8.4% for the single measurements and 8.0% for the mean of triplicate measurements. The correlations between the first measurements and the mean of triplicate measurements ranged from 0.90 to 0.98 (all p less than 0.01). The average difference between the two visits for all four blood pressure parameters was -0.6 mm Hg for the single measurements and -0.5 mm Hg for the mean of triplicate measurements (all p = NS). These results indicate that 1) blood pressure does not change further after 1 week of placebo treatment, and 2) use of the mean of triplicate measurements of blood pressure and heart rate gives the same result as use of single measurements, and the results are no less variable.