A subject, who had undergone surgery to replace one hip joint and the proximal half of the femur with an instrumented titanium implant, performed brief exercises whilst simultaneous measurements were made of compressive axial force in the implant using short-range wireless telemetry, ground reactions using a Kistler force plate, and electromyographic activity of the vastus lateralis (VL) and erector spinae (ES) muscles using surface electrodes. Recordings were made barefoot and wearing 'trainers'. The exercises (slow jumping in counter movement style, fast continuous jumping, and jogging on the spot) have been found effective in controlled interventions for increasing bone mineral density in women. The implant forces were 250-400% BW. The values were about twice the magnitude of the ground reaction forces and significantly correlated with them for both peak force and its rate of rise but their relative magnitudes varied depending on mode of activity (jumping or jogging). Implant forces were significantly related to the muscle activity; in multiple regression analysis implant forces during take off from slow jumps VL contributed significantly in addition to the ground reaction (98% total explained variance). There was more activity in VL during jumping than jogging for the same implant force which may explain why jumping appears to be more osteogenic than jogging for the femur. For the same ground reaction, wearing trainers increased both the magnitude of the compressive loading of the femur and its rate of rise.