Because the correspondence between laboratory measures of blood pressure and heart rate responses to stress and ambulatory measures is less than optimal, this study tested two hypotheses: Are ambulatory measures of blood pressure elevated during periods of perceived stress, relative to no stress? Are ambulatory blood pressures elevated during perceived stress among those individuals who exhibit elevated blood pressure and heart rate responses to laboratory stress? These questions were addressed in a sample of employed, middle-aged men and premenopausal and postmenopausal women, who vary in reproductive hormone status, and in risk for coronary heart disease. All participants performed a series of laboratory studies while their physiological parameters were monitored and then wore an ambulatory blood pressure monitor for a day and a half. This monitor recorded blood pressure every half hour during the waking hours and at the same time the participants assessed their mood states. After excluding participants who reported no variability in stress levels, those who were cardiovascular reactors to a laboratory speech task exhibited elevated ambulatory blood pressure levels during periods of perceived stress. Furthermore, in general, periods of perceived stress were associated on a within subject basis with elevated ambulatory blood pressure. These results suggest that the correspondence between laboratory and field measures of blood pressure would be improved by taking into account the environmental circumstances during the ambulatory assessments and the person characteristics of reactor-nonreactor to laboratory stress.