Five hundred and eleven McKee-Farrar prostheses implanted at the Orthopedic Hospital of the Invalid Foundation (Helsinki, Finland) between 1967 and 1973 were analyzed for long-term survivorship, clinical results, and the risk factors leading to aseptic loosening of the prosthesis. Reoperation was necessary for aseptic loosening of 91 prostheses and for septic loosening of 19 prosthesis. Ten-year survivorship was 76% and the annual failure rate 2.3%. The failure rate correlated with time. Acetabular cup loosening occurred 1.2 times more often than stem loosening. The retained prostheses remained clinically stable with respect to pain and mobility, but some deterioration was observed in walking ability. In addition to the follow-up period, prosthetic design, right-sided prosthesis, and lengthening of the operated leg proved to be significant factors for aseptic loosening in multiple stepwise regression analysis. After exclusion of the unmeasured leg length inequalities, the lengthening of the operated leg proved to be the most important factor explaining aseptic loosening of the prosthesis. Surgeon's inexperience, operation time, post-traumatic coxarthrosis, and previous operations were correlated to aseptic loosening of the prosthesis.