This study was designed to compare the biomechanical and functional characteristics of allograft/endoprosthetic composites of the proximal 25% of the femur repaired with either a transverse or a step-cut osteotomy, using a canine model (10 dogs, five with each type of osteotomy). Serial radiography and weight-bearing studies were performed monthly, and mechanical testing was done 6 months after surgery. The femora were tested in torsion and compared with the contralateral control (insertion of a femoral component but no osteotomy). At 6 months, the composites with a step-cut osteotomy had 36% greater structural stiffness than the composites with a transverse osteotomy (p < 0.005) and 121% greater maximum torque at failure than the controls (p < 0.005), without greater structural stiffness. Evaluation of peak vertical ground reaction forces revealed significantly greater weight-bearing on the experimental limb in dogs with a transverse osteotomy. The results of this relatively short-term study were mixed. Despite the increased structural stiffness of the allograft/endoprosthetic composite with a step-cut osteotomy, the dogs with this type of reconstruction had decreased weight-bearing throughout the course of the study. The step-cut osteotomy may augment the stability of the allograft/endoprosthetic composite, allowing faster healing (as demonstrated by the results of mechanical testing), but in some way, not understood, may cause pain in the reconstructed limb. Longer term studies are needed to answer these questions and to determine whether alteration of the traditional transverse osteotomy has any advantage.