We reviewed the results of forty-two total hip replacements that had been done with the self-locking Autophor ceramic total hip prosthesis in thirty-four patients. The patients ranged in age from twenty-five to sixty-seven years (average, forty-eight years). The diagnoses were avascular necrosis (eight patients), degenerative arthritis (nineteen patients), ankylosing spondylitis (one patient), post-traumatic arthritis (two patients), and rheumatoid arthritis (four patients). Seven procedures were revisions: five, of a loose cemented total hip prosthesis and two, of a loose noncemented endoprosthesis. No revision was done for infection. The length of follow-up ranged from twenty-seven to sixty-six months (average, fifty-one months). The patients were evaluated by physical examination, serial radiographs, and questionnaires. The hips were rated with the modified Harris hip score. At the time of follow-up, eleven of the twenty-seven patients who had had a primary hip replacement complained of at least moderate pain that limited activities; however, only three patients had to have a revision. The femoral components had a notable tendency to subside more than five millimeters, and in fifteen hips there was radiographic evidence of progressive loosening. Our experience with the self-locking Autophor ceramic total hip prosthesis has been disappointing. We no longer use it.