In 20 male subjects with effort angina, an ambulatory ECG monitoring (AEM) was registered and an exercise test was carried out immediately afterwards. Usual indications for termination of test have been modified to detect the temporal relationship between the appearance of ischemic ST changes and appearance of pain. During AEM 15 patients (pts) presented ischemic episodes with a total of 80 episodes, of which 46 were asymptomatic and 34 accompanied by pain. Among the ischemic episodes registered by AEM, the mean duration of asymptomatic episodes was less that of symptomatic ones. The same can be said for the mean entity of maximum ST depression. On the basis of the temporal relationship between the appearance of ischemic ECG features and pain during exercise test, we classified 4 groups of patients: 1) pts with an almost contemporary appearance of ST alterations and pain (5 pts); 2) pts where pain appeared with a delay variable between 10 and 30 sec (6 pts); 3) pts where pain appeared with a delay of more than 30 sec (4 pts); 4) pts with absence of pain (5 pts). In these pts the exercise test was stopped either because ST depression had reached 3 mm or because ST alterations persisted for 3' without pain. Comparing the responses of the exercise test with the data of AEM, a statistical relationship was found between patients from the different groups and the prevalence of asymptomatic or symptomatic episodes of ischemia. Asymptomatic ischemic episodes during AEM are more frequent in pts who during exercise test show ischemic ECG features without pain or in pts where pain appears with a noticeable delay in comparison to ECG ischemic alterations.