The effects of locking screw position (long column fixation--long distance between the nearest screws to the fracture--and short column fixation--short distance between the nearest screws to the fracture) and fracture gap size (1-mm and 8-mm transverse fracture gap) on stiffness and fatigue of fractured femur fixed with a locking compression plate (LCP) were biomechanically evaluated. The stiffness of 1-mm fracture gap models and that of intact femoral model were in the range of 270-284 N/mm, while those of 8-mm fracture gap models were significantly lower (155-170 N/mm). After 1,000,000 cycles of loading, no fracture of LCP of 1-mm fracture gap models fixed in either long column or short column fashions occurred. On the other hand, the complete fractures of LCPs of 8-mm fracture gap models fixed in long column and short column fashions occurred after 51,500 and 42,000 cycles of loading, respectively. These results suggest that the full weight loading may be allowed for the patient with 1-mm transverse femoral fracture fixed with an LCP. On the other hand, the full load of walking should be avoided for the patient with 8-mm transverse femoral fracture fixed with an LCP before adequate healing.