The relationships among angina pectoris, stress tests, and arteriography are complex. The majority of patients with angina pectoris can be adequately diagnosed by a careful history. Considerable attention to detail and repeated questioning is often necessary before the pain syndrome can be accurately classified. The resting electrocardiogram is of limited value in the diagnosis despite the fact that there is a high positive correlation between abnormal ST-T changes on the electrogram and significant obstructive lesions on coronary arteriograms. The value of the electrocardiogram is enhanced, and its specifiicity and sensitivity increased, when used in combination with exercise stress. The lowest error percentage is achieved by utilizing rate-standardized exercise tests and multiple leads with loads that produce heart rate responses of 80-90% of the expected maximum. Coronary arteriography gives the most specific anatomic information in patients with ischemic cardiac pain but will not directly disclose the cause of the pain. This fact assumes considerable importance when the pain has atypical features or when the patient is in the age group that has a high prevalence of coronary atherosclerosis. Similarly, the presence of past myocardial infarction is likely to be associated with obstructive disease, regardless of the cause of the patient's current symptoms. The exact role of lipid and other metabolic abnormalities in producing coronary arteriographic changes in the absence of symptoms needs further clarification, although the available data suggest that marked elevations in lipid fractions are frequently associated with atherosclerotic change, regardless of symptoms. Finally, the data imply that the anatomic abnormalities and functional consequences of the coronary atherosclerotic process are more important predictors of the patient's course than any specific symptomatic expression such as angina pectoris.