Twenty-one patients who had primary osteoarthrosis were managed with a bilateral total hip arthroplasty with insertion of a Charnley femoral component. In each patient, one hip was randomly allocated to have a Harris-Galante acetabular component inserted without cement, and the contralateral hip was treated with an all-polyethylene Charnley acetabular component fixed with cement. The clinical result was satisfactory in all patients. All forty-two hips were followed, with respect to migration of the acetabular component, with use of roentgenstereophotogrammetric analysis for a median of twenty-seven months (range, twenty-three to forty-nine months). Each patient served as his or her own control. Maximum migration in any direction was 1.7 and 2.1 millimeters, and maximum rotation was 2.2 and 2.0 degrees for the Harris-Galante and Charnley acetabular components, respectively. There was no significant difference in migration between the two designs of acetabular components (p = 0.98, p = 0.75, and p = 0.06 for the transverse, longitudinal, and sagittal axes, respectively). However, the Harris-Galante acetabular components rotated significantly more than the Charnley acetabular components around two of the three axes (p = 0.008, p = 0.08, and p = 0.03 for the transverse, longitudinal, and sagittal axes, respectively). The Charnley hip implant has been used clinically for a long time, with successful results. Comparison of new designs of implants with the Charnley prosthesis is therefore important. Roentgenstereophotogrammetric analysis provides a potential for detection of problems with fixation at an early stage rather than after long-term follow-up. No major difference in terms of skeletal fixation was found between the two designs of components after short to medium-term follow-up.