The distribution of electrocardiographic (ECG) abnormalities and the relationship between ECG abnormalities and mortality after 20 years of follow-up were examined among 40-64-year-old black men and white men enrolled in the Evans County Heart Study. Major or minor ECG abnormalities, as defined in the Pooling Project, were present at entry for 53% of blacks (164 of 308) and 31% of whites (159 of 511). For both races, the presence of ECG abnormalities was directly related to age, blood pressure, and Quetelet's index at baseline. After adjustment for these and other risk factors, major ECG abnormalities were similarly predictive of all-cause mortality for blacks [rate ratio (RR) = 1.7 (126.96.36.199)] and whites [RR = 2.2 (1.4, 3.4)]. Associations of similar magnitude were observed in relationship to deaths from all cardiovascular diseases and deaths from coronary heart disease. These results are consistent with the hypothesis that ECG abnormalities convey risk for blacks as well as whites.