Two hundred and four primary Mayo total ankle arthroplasties were performed in 179 patients at the Mayo Clinic from 1974 through 1988. We evaluated the clinical result after 160 arthroplasties in 143 patients who had been followed for two years or more (mean, nine years; range, two to seventeen years). The result was good for thirty-one ankles (19 per cent), fair for fifty-five (34 per cent), and poor for seventeen (11 per cent); fifty-seven arthroplasties (36 per cent) were considered to be a failure (defined as removal of the implant). Adequate preoperative and follow-up radiographs were available for 101 ankles (eighty-nine patients). There was radiographic evidence of loosening of eight tibial components (8 per cent) and fifty-eight talar components (57 per cent), but we found no association between the clinical and radiographic results. Complications occurred after nineteen (12 per cent) of the 160 arthroplasties, and ninety-four additional reoperations were necessary after sixty-six (41 per cent).On the basis of these findings, we do not recommend ankle arthroplasty with a constrained Mayo implant for rheumatoid arthritis or osteoarthrosis of the ankle.