In a seven-year follow-up of 3,102 Evans County, Georgia, residents examined in the prevalence survey of 1960 through 1962, 143 new cases of coronary heart disease (CHD) occurred, 56 of which resulted in death during the interval. Ethnic differences in CHD among males discovered during 1960 through 1962 were again confirmed.Excess of white males could not be explained by differences in case fatality, diagnostic criteria, missed cases, or by a competing cause of death. In contrast to ethnic differences, differences by social class found in the prevalence survey among white males were not found in the incidence study. Various analyses suggested that the excess prevalence found in the high social class was a reflection of prior high incidence rates in this class. Over time these class differences are disappearing, particularly among younger men.As indicated previously,1 the prev-. alence study in Evans County, Georgia, in 1960 through 1962 had shown marked differences in cor¬ onary heart disease (CHD) by eth¬ nic group and social class. White males had an age-adjusted prevalence rate that was almost three times higher than black males (59 per 1,000 compared to 22 per 1,000). Most of this white excess was attributed to the higher social class men who had a rate of 99 compared to 40 for the lower social class. Further, no differ¬ ences were found by ethnic group for females, the rate for white females being 18 and for black females 14 per 1,000. Thus, within ethnic groups, white males had a five-fold higher rate than white females, while for blacks the ratio was only 2 to 1.The initial purpose of the incidence study was to determine whether these differences persisted when a population free of CHD was followed over time. Such an analysis would provide an indication of the extent to which the differences in prevalence in the various groups indicated differ¬ ences in risk of developing CHD as opposed to differences in selective survival or migration, or both. Fur¬ ther, it was hoped that this follow-up study might indicate whether, in the rapidly changing environment of Evans County, a secular trend in these ethnic group-social class pat¬ terns was occurring.The methods by which this study was conducted, including the com¬ pleteness of ascertainment and ex¬ amination of the members of the original 1960 through 1962 cohort and the criteria used for diagnosis of the various manifestations of CHD, are described elsewhere.2 Results Differences by Ethnic Group andSex.-The incidence of all manifesta¬ tions of definite and probable CHD for the 87-month follow-up period is shown by age, ethnic group, and sex in Table 1 and Fig 1.As can be seen from these data, the differences in males by ethnic group found in the prevalence study were confirmed, black males having lower rates at all ages than white males, with only minor differences appear¬ ing among females. The extraordi¬ narily low rates for black males in comparison to the whites was one of the most intriguing findings of this study and one for which explanations were energ...