A population of 3983 apparently healthy people whose mean age at entry was 27.0 years was followed with periodic medical examinations and ecg''s for 10.6 years. Based upon the date of last contact the mean exposure was 11.9 years. There were 83 deaths, of which 7 were from myocardial disease. Thirty-two subjects developed clinical myocardial disease at a mean age of 48 years. Of those who entered the study with abnormal ecg''s, 3.3% have so far developed myocardial disease in contrast with 0.7% for those whose initial ecg''s were normal. Some (280) subjects 7% of the population, on at least one occasion, had an abnormal ecg. Partial heart block, paroxysmal tachycardia, shifts in S-T level and primary T-wave changes tended to occur intermittently, while QRS voltage changes, pathologic Q waves, complete heart block, Wolff-Parkinson-White configuration, and bundle-branch block usually remained constant once they appeared. At the observed ages and durations the subjects whose ecg''s exhibited partial heart block, Wolff-Parkinson-White configuration, and right bundle-branch block have remained, with one exception, in apparent good health. Low QRS voltage, high QRS voltage, and primary T-wave changes with or without a shift in the S-T segments were the only abnormal ecg components that preceded the onset of myocardial disease but many showing these and other abnormalities in the ecg have not so far developed clinical evidence of disease. More attention to technical details and standardization requirements would reduce the number of abnormal ecg''s in apparently healthy people. This long-term follow-up study, the object of which was to determine the clinical significance of abnormal ecg''s in apparently healthy people, has now a well-established baseline on a population of 3983 persons. As the majority of those under investigation are approaching the age at which myocardial disease becomes clinically recognizable, the value of the study will increase in the years immediately ahead.