We used dual-energy X-ray absorptiometry (DEXA) to measure the bone mineral content (BMC) of both tibiae in 13 patients who had been treated for a tibial fracture by rigid plate fixation. Within two weeks of plate removal the BMC was significantly greater in the bone that had been under the plate than at the same site in the control tibia. An unplated area of bone near the ankle showed a significant decrease in BMC at the time of plate removal with subsequent return to the level of the control tibia during the ensuing 18 months. We conclude that osteoinductive influences outweigh the potential causes of osteopenia, such as stress shielding and disuse, and that, contrary to expectation, demineralisation is not a factor in the diminished strength of the tibia after plating for fracture.