Systolic time intervals (STI) and apex cardiographic A wave amplitude were measured to assess their predictive value for coronary heart disease in 1231 men aged from 17 to 64 years. Thirteen men had had a previous myocardial infarction (MI) and during a 4-year follow-up there were 30 new MI cases. Men with previous MI showed prolongations of the STI, but when the STI were corrected for the effect of age, heart rate, blood pressure and body size, the differences did not remain statistically significant. On the other hand, men who had MI during the follow-up, had significantly prolonged STI at the initial examination, and this difference remained significant also after correction for the other factors. The apexcardiographic A wave did not show any consistent relationship either to the presence of old MI or the occurrence of a new MI. Analysis of the incidence of new MI by quintiles of STI showed that the 4-year incidence of MI was related to the duration of electromechanical systole (QS2) and pre-ejection period (PEP). 80% of new MI cases were in the two highest quintiles of QS2 and MI incidence increased gradually from 0.7% in the quintile with shortest PEP to 6.9% in the highest quintile of PEP. Multiple regression analyses revealed that the duration of QS2 or PEP had a significant, positive correlation with the incidence of MI, even when the effect of age and blood pressure was taken into account. In conclusion, myocardial dysfunction detectable by means of mechanocardiography in asymptomatic subjects has a predictive value with respect to future myocardial infarction.