The prevalence of femoral osteolysis in hips in which a femoral component had been inserted without cement was compared with that in hips with a cemented component, in a retrospective matched-pair study of the results of primary total hip arthroplasties; all patients had received the same type of acetabular component. Forty-one hips in thirty-nine patients who had a Harris-Galante porous-coated total hip prosthesis without cement were matched by age, sex, weight, duration of follow-up, and diagnosis with forty-one hips in thirty-eight patients who had a hybrid total hip reconstruction; the hybrid reconstruction consisted of the same acetabular component and a Precoat femoral component inserted with a so-called third-generation cementing technique. All of the operations were done by the same surgeon, who used the same operative approach and the same course of postoperative rehabilitation. All of the patients were followed for at least four years (mean, six years). Osteolysis developed in twelve (29 percent) of the hips that had a femoral component without cement compared with none of the hips that had a cemented component (p < 0.0002). At the latest follow-up examination, none of the femoral components that had been inserted with cement were loose and none had been revised, while eight (20 percent) of the femoral components that had been inserted without cement were loose and five (12 per cent) had been revised. This retrospective matched-pair study controlled for many of the variables associated with a comparison of the rates of femoral osteolysis in separate series of femoral components fixed with and without cement.(ABSTRACT TRUNCATED AT 250 WORDS)