First, to test whether there was a change in the blood pressure and heart rate on work days relative to non-work days; secondly, to assess whether there were work-home differences on these days; and, thirdly, to assess whether these changes were similar in both normotensives and hypertensives.Twenty-four working men (17 mild hypertensives and seven normotensives) wore an ambulatory blood pressure monitor for two 24-h periods (at work and home), on a work day ('on day') and on a non-work day ('off day').Means were calculated for each day for each home and work period. On the off day, the 'work' mean was defined using the readings during the usual working hours. To control for the effect of position on the means, statistical procedures that control for position were employed. These means were then analyzed by repeated-measures analysis of variance with two within-person factors. Period (home-work) and Day (on-off), and one between-person factor, Hypertensive Status (mild hypertensive-normotensive).When hypertensives and normotensives were combined, diastolic blood pressure (DBP) was significantly greater at work than at home. However, for both systolic blood pressure (SBP) and DBP this work-home difference was dependent on whether it was the on or the off day, the work-home difference being greater on the work day. Furthermore, the noted work-non-work day changes between work and home were significantly different for DBP for normotensives and mild hypertensives. Normotensives experienced a 6-mmHg decrease in blood pressure on the off day, whereas mild hypertensives had a 3-mmHg average increase on the off day. No significant differences were found in the pattern of work-home differences between on and off days for mild hypertensives and normotensives, although the effect approached significance for SBP, providing an indication that the work-home differences may depend on both hypertensive status and work-non-work day. Heart rate was significantly higher during work than at home. Although normotensives had higher heart rates, the work-home difference was constant across on and off days. Among 10 subjects who agreed to wear the monitor a third time on a second work day, the average work blood pressures during the two work periods were nearly equivalent, arguing against an habituation effect to the monitor.These data support the notion that, for blood pressure, activity is an important determinant of level. A circadian basis to blood pressure was not supported. We found differences between work and non-work days for normotensives, and a trend towards differences by work-home period and work-non-work day for hypertensives. For heart rate the opposite result was found. Heart rate was affected only by time of day (as denoted by work and home periods) and was stable across on and off days. These results suggest that experimental or statistical control of activity and position may be necessary if ambulatory blood pressure monitoring is to be used as either a diagnostic or research tool.