Introduction. The French paradox regarding cemented femoral components has not been resolved, so we compared the mechanical behavior of a French stem, the CMK stem (Biomet, Warsaw, IN, USA), with a collarless, polished, tapered stem (CPT, Zimmer, Warsaw, IN, USA) using an original biomechanical instrument. Materials and Methods. Two size-3 CPT stems and two size-302 CMK stems stems were fixed with bone cement into a composite femur soaked in vegetable oil to simulate wet condition. The composite femur was attached to a biomechanical testing instrument after stem implantation, and a 1-Hz dynamic sine wave load (3000 N) was applied to the stems for a total of 1 million cycles. An 8-hour unload period was set after every 16 hours of load. Femur temperature was maintained at 37°C during testing. The femoral canal was prepared for the CPT stems by standard rasping; for the CMK stems, however, the French method was used, in which cancellous bone was removed with a reamer. One CMK stem (CMK-1) was inserted into a femur without collar contact (>2 mm above the calcar), and the other (CMK-2) was inserted into a femur with collar contact. Stem subsidence was measured at the stem shoulder. Compressive force and horizontal cement movement were measured via rods set at the cement–bone interface on the medial, lateral, anterior, and posterior sides of the proximal and distal portions of the composite femurs. Results. Subsidence was as follows: 0.521 mm and 0.629 mm for the CPT stems, 0.46 mm for CMK-1, and 0.36 mm for CMK-2. Compressive force at the cement–bone interface was at the maximum level at the proximomedial portion of all stems. These forces increased gradually until the one-millionth loading. Maximum compressive forces were 183 N and 107 N for the CPT stems, 180 N for CMK-1, and 215 N for CMK-2. There was a strong positive correlation between stem subsidence and compressive force in all stems. Radial cement creep at the proximomedial portion was 90 μ for one of the CPT stems, 184 μ for CMK-1, and −636 μ for CMK-2. Discussion. We previously reported our findings of a positive correlation between stem subsidence and compressive force in CPT stems. In the current study, CMK stems also subsided even when there was stem collar contact with bone. Subsidence was less in CMK stems than in CPT stems, but the values were close. In addition, compressive force and radial cement creep in CMK stems were also similar to or greater than in CPT stems. Conclusion. The two different concept stems demonstrated similar behavior in relation to bone cement, a finding that may present a solution to the French paradox