Wireless telemetry of forces from within the human body was applied to measurements of the axial force in a Harrington rod used for internal correction of scoliosis. The heat sterilizable system contains a force-gauging pressductor, an internal power transfer coil, and an internal transmitter. The electrical energy for the transmitter and pressductor is provided by means of electromagnetic induction between an external power transfer coil and the implanted one. The implanted units were covered with epoxy resin and silicone rubber and the wires were Teflon-coated. The modified Harrington rods were inserted in four girls, aged twelve to sixteen years, who had idiopathic scoliosis of varying severity (55 to 90 degrees). At a second procedure two to three weeks after the first (a routine procedure in our department) the rod was replaced by a conventional one, the implanted transmitter and pressductor were removed, more correction was obtained, and spine fusion was performed. In none of the patients was there any evidence of harmful effects from the implanted devices, either clinically or morphologically. Distracting forces during operation which ranged from twenty to forty kiloponds resulted in corrections of 55 to 70 per cent. None of the curves was particularly rigid. The force declined with time, rapidly in the beginning and slower after three days. After about ten days the recorded force stabilized at about one-third of the maximum force used during operation. Recordings demonstrated the need for great care in handling the patients immediately after operation to avoid excessive loading and fracture of the lamina under which the upper hook was inserted. The axial forces during deep breathing, bicycling in bed, and similar exercises were relatively low. Log-rolling to one side caused only a slight increase which was less when the patient was lying on the side of the concavity of the curve than on the opposite side. In the standing position the Milwaukee brace, compared with no external support, exerted a distractive force of four to six kiloponds in the three patients tested. In the supine position, compared with no external support, the brace caused a reduction in axial force of two to four kiloponds. Evaluation of the effect of the different parts of a Milwaukee brace in the standing position demonstrated the importance of a well fitted pelvic cage, of either the chin or the occiput pad, and especially of two side-supporting pads, one on each side. The results with the Milwaukee brace should be valid for patients with moderate curves of idiopathic origin, for whom this brace is commonly used without operation.