The long-term results of aortic valve replacement with the fresh aortic homograft, performed in 114 patients at Stanford University Medical Center from 1967 to 1971, were evaluated. There were 10 operative deaths (8.8 per cent), only 3 (5 per cent) in the period from 1968 to 1971. There were 6 late deaths in the first year (5.8 per cent) and 8 in later years (1.5 per cent per year); 12 late deaths were due to cardiac causes, 6 of them to valve dysfunction. The homograft was replaced later with a prosthetic valve or heterograft in 22 patients (3.2 per cent per year): for regurgitation in 20 and for calcific stenosis in only one. Infective endocarditis occurred in 5 cases, accounting for one operative death, 2 late deaths, and 2 reoperations with survival. Systemic thromboembolism occurred in 6 patients, 3 with mitral valve disease, one with atrial fibrillation, and one with infective endocarditis; none was a proved instance of embolism from bland thrombus on the aortic homograft valve. Of 53 patients followed for 5 years or more with the homograft intact, 47 have minimal or no disability, despite aortic diastolic murmurs in many. We conclude that long-term results are good in the majority of patients, with aortic regurgitation requiring reoperation being the leading complication. These results may serve as a basis for comparison of more recently introduced methods of aortic valve replacement.