Increased platelet aggregation in the morning and upon assuming an upright posture may account at least in part for the observed circadian variation in onset of acute myocardial infarction. The Physicians' Health Study, a randomized, double-blind, placebo-controlled trial of alternate-day aspirin intake (325 mg) among 22,071 US male physicians, afforded the opportunity to assess this circadian pattern and examine whether it is altered by aspirin therapy. During a 5-year period of follow-up, 342 cases of nonfatal myocardial infarction were confirmed, of which the time of onset was available in 211 (62%). The placebo group showed a bimodal circadian variation in onset of myocardial infarction with a primary peak between 4:00 AM and 10:00 AM (p less than 0.001). In the aspirin group, however, this circadian variation was minimal (p = 0.16), due primarily to a marked reduction in the morning peak of infarction. Specifically, aspirin was associated with a 59.3% reduction in the incidence of infarction during the morning waking hours, compared with a 34.1% reduction for the remaining hours of the day. The greater reduction was observed during the 3-hour interval immediately after awakening, a period with a risk of infarction twice that of any other comparable time interval (p less than 0.001). Aspirin intake was associated with a mean reduction in the incidence of infarction of 44.8% over the entire 24-hour cycle. These data support the hypothesis that increased platelet aggregability in the morning and upon arising contributes to the occurrence of myocardial infarction and that aspirin reduces the risk of infarction by inhibiting platelet aggregation during these critical periods.