A retrospective analysis of data from 207 non-pregnant premenopausal women showed that the mean level of systolic blood pressure varied with the stage of the menstrual cycle, being higher on days 17-26, the part of the luteal phase during which the peak of progestogen levels develops, than during the luteal phase as a whole, and significantly higher than the mean for all other days of the cycle. The mean levels were 125.4 mmHg (SE 1.76) for days 17-26, 122.5 (SE 1.25) for days 15-28 and 120.1 (SE 1.07) for days 1-16, 27 and 28. This finding supported the hypothesis that endogenous progestogen might have a hypertensive effect, as does exogenous progestogen. However, a second study designed to confirm this finding failed to do so, showing no cyclical change in the level of blood pressure. The subjects in the first study may have been subject to greater psychological stress when the measurements were made than were those in the second, a possibility supported by the large difference in pressure between the two studies. The discrepancy between the two sets of results could be explained if the effect possibly associated with progestogen levels in the retrospective study was due not to a hypertensive action per se but to a progestogen-related increased reaction to stress.