From July 1984 through October 1987, 63 consecutive primary Porous Coated Anatomic total hip arthroplasties were performed. Follow-up observation ranged from one to four years, with a mean of 2.4 years. The mean postoperative Harris hip score in 81 cases was 38% excellent results, 23% good results, 6% fair results, and 23% poor results. Eight percent of hips required revision of the femoral component related to subsidence. Persistent thigh pain was found in 33% and persistent limp in 58%. Poor clinical outcome was correlated with persistent thigh pain and 3 mm or more subsidence of the femoral component. Subsidence occurred despite adequate canal fill. This high percentage of early failure is unacceptable when compared with standard techniques of total hip replacements with cement fixation.