Clinical and radiographic evaluations of 15 McKee-Farrar hip replacements in 13 patients with followup of 21 to 26 years were performed. The average Harris hip score was 86 with no patients having a poor result. These patients outscored the age matched controls in all categories of the SF-36 health survey. All patients were community ambulators with qualitative activity levels exceeding the average for their age. Quantitative activity assessment with a pedometer in 3 patients indicated a current average of approximately 900,000 cycles per year. This represents more than 21 million cycles when extrapolated during the life of the implants. None of the femoral components were radiographically loose. One acetabular component may be loose. Osteolysis developed in 3 apparently well fixed femurs and in 1 acetabulum. There were several features of these cases that may have contributed to the long survival: (1) relatively small stature of the patients who averaged 160.5 cm (5 feet 5 inches) in height and 66.9 kg (147 lbs) in weight; (2) favorable biomechanics of the reconstruction with the hip center of rotation being medialized by an average of 6.4 mm and the femoral offset increased by an average of 4.9 mm; (3) decreased potential for neck socket impingement with an average lateral acetabular opening of 54 degrees and all components were anteverted; (4) radiolucent cement in 13 of 15 hips; and (5) no radiographically measurable wear. Previous analyses and comparisons of the clinical performance of the McKee-Farrar implant have focused on the metal on metal bearing. As has been recognized with the many variations of total hip replacement using metal on plastic hearings, there are a myriad of variables that contribute to clinical outcome. The results of this study suggest that patient selection and technical factors may contribute to the long term survival, and conversely to the failure, of McKee-Farrar implants.