Data concerning the time of onset of myocardial infarction were obtained for 540 of the 544 patients with creatinine kinase (CK)-MB-confirmed non-Q wave myocardial infarction enrolled in the multicenter Diltiazem Reinfarction Study. Data were also collected for 627 patients who were screened but excluded. Among the 1,167 patients, no diurnal pattern of onset could be found at either 2- or 6-hour intervals. Among the 540 patients enrolled in the trial, no pattern could be found at these intervals either, although at 8-hour intervals, 27% of infarctions occurred between midnight and 8:00 AM, compared with 37% between 8:00 AM and 4:00 PM and 36% between 4:00 PM and 12:00 AM (p = 0.02). In contrast to the patterns previously noted for Q wave myocardial infarction, there was no preponderance of non-Q wave infarction in the late morning. Circadian rhythm was also absent among patients not treated with beta-blockers as well as among patients presenting with ST segment elevation on their enrollment electrocardiograms. Diabetics, women, and patients with first infarction were more likely to present during the afternoon hours. We conclude that the late morning preponderance seen for Q-wave myocardial infarction is not discernable in patients with non-Q wave myocardial infarction. This observation suggests that the pathogenesis of these two infarct subtypes is different or that the process of thrombotic coronary occlusion in Q wave infarction (sustained) differs from that in non-Q wave infarction (nonsustained).