Total joint replacement patients today are younger, heavier, and more active than total joint replacement patients 40 yrs ago. Consequently, patient expectations and prosthesis requirements have increased and there is a need to re-evaluate preclinical testing methods. We present the design rationale for a novel load simulator for the proximal femur, capable of applying a more aggressive load profile than previous simulators. This simulator was used to measure three-dimensional micromotion of a cemented total hip replacement femoral stem under simulated physiological loading. We assessed the influence of a separate abductor muscle force, a higher joint reaction force, and a more accurate implant stability measurement system included in the new simulator and compared the results to the lower, single joint reaction force included in a previously published simulator. Per-cycle motion at both cement interfaces and stem and cement mantle migration obtained from both simulators using the same femoral stem design, are compared. Although the new simulator applied higher loads, per-cycle motions were lower than previously reported. In both studies, regardless of the presence or lack of a separate muscle force, the greatest motions were in the medial-lateral direction (new: 27 +/- 4 mum, old: 67 +/- 21 mum). The findings indicate that magnitude and direction of peak joint reaction force and inclusion of a separate muscle force have a significant effect on femoral stem stability measurements. We recommend that future femoral stem stability studies consider using load simulation techniques and a direct motion measurement system comparable to the one presented in this study.