The phases of systole were measured in 51 patients with acute myocardial infarction and three control groups: (1) a group of 40 patients without heart disease, (2) a group of 23 patients admitted to a coronary care unit for chest pain, who did not have an acute myocardial infarction, and (3) a group of 16 patients with stable angina pectoris and arteriographically proven coronary atherosclerosis. In addition, serial measurements were made in the acute myocardial infarction group. Total electricalmechanical systole (QS2), the preejection period (PEP), and left ventricular ejection time (LVET) were measured in each patient from simultaneous recordings of the ECG, phonocardiogram, and carotid pulse tracing. The systolic and diastolic blood pressures and QRS duration were also measured. Corrections were made for heart rate where appropriate.The average PEP was elevated on the first day of myocardial infarction but was within normal limits thereafter. The LVET and QS2were significantly shortened until the fourth week of hospitalization. There was considerable overlap in the PEP values among the four groups. Eight patients died of acute myocardial infarction; the PEP was abnormally short in three (two of whom had cardiogenic shock), normal in two, and abnormally long in three. The PEP/LVET ratio separated the acute MI group from the normal group but not from the other two patient groups. Clinical class, digitalis, and infarct location did not produce characteristic changes in the systolic time intervals.A reduction in stroke volume is the most likely explanation for the reduction in LVET and QS2. The wide range in PEP values observed is best explained by alterations in the multiple determinants of PEP.The systolic time intervals do not appear to be useful as a diagnostic or prognostic tool in acute myocardial infarction.