There is confusion in the literature as to the effect of ageing on blood pressure variability. The lack of consistency in reports probably reflects differences in blood pressure measurement techniques and in the choice of statistics used to describe variability. We studied 16 hypertensives (clinical blood pressure greater than 140/90 mmHg on 3 occasions) over 60 (67 +/- 5.5 years) and 16 under 60 years of age (44 +/- 8.9 years) using the Remler M 2000 ambulatory system to measure blood pressure and heart rate every 30 min during the awake hours of the day. Mean +/- s.e.m. blood pressure and heart rate for the elderly was 168 +/- 2.5/95 +/- 1.0 mmHg and 72 +/- 1.5 beats/min, corresponding values for the young being 162 +/- 5.3/103 +/- 2.1 and 78 +/- 2.5. Four statistics of variability were used: standard deviation, coefficient of variation, range and mean hourly change. Differences between old and young were found only for the range of systolic and diastolic blood pressure which was lower in the elderly (52 +/- 3.9/32 +/- 2.4 versus 67 +/- 5.2/46 +/- 4.8 mmHg, both P less than 0.05) and mean hourly change of heart rate (P less than 0.05). The standard deviation of heart rate was negatively correlated with age (r = -0.37; P less than 0.05). Only the systolic and diastolic blood pressure ranges were found to differ with age; the more rigorous tests of variability, standard deviation and coefficient of variation, were not changed. These data show that in hypertensives of different age but with roughly comparable blood pressure levels, variability of blood pressure is not consistently related to age.(ABSTRACT TRUNCATED AT 250 WORDS)