AbstractAs is true for laboratory or clinical research, epidemiology can either generate or test hypotheses about the origins of disease. In so doing with respect to cardiac anomalies, rubella virus was identified as an environmental cause and the role of genetics has been clarified. Microscopically visible chromosomal aberrations have been related to high rates of diverse cardiac defects in Turner's and autosomaltrisomy syndromes. Submicroscopic genetic defects may also produce heart anomalies, as in Marfan's syndrome. The role of inheritance in the production of cardiac malformations in general, however, appears to be small as indicated by studies of twins (only 1 of 6 series revealed an elevated concordance rate in identical pairs) or by studies of the effect of consanguinity. Yet a wide variety of congenital heart disease has been repeatedly found to occur excessively in families. Future research must evaluate whether these occurrences are determined pre‐ or postzygotically. By so doing the origins of sporadic cases may be elicited. The possibilities include submicroscopic chromosomal abnormalities, chemical or microbial agents that interfere with normal embryonic development, and combinations of these factors. Through the use of existing data resources epidemiologic research can help to identify the influence of each. Although computer technology has its value, etiologic insight, which can be tested by the epidemiologic method, seems more likely to come from observations made by physicians at the bedsid