&NA;Hyperthermia has significant potential as an adjuvant form of brain tumor therapy. Current intracranial hyperthermia methods, however, are limited in their ability to control spatiotemporal thermal distribution. A stereotaxic magnetic movement system that may be capable of heating discrete regions of brain to a preselected temperature is described. With this system, a ferromagnetic object (referred to as a thermoceptor) is directed through the brain by an external drive magnet. Real time thermoceptor position is monitored with biplanar fluoroscopy and superimposed on a preoperative magnetic resonance imaging scan using a computer. Once in position, the thermoceptor can be inductively heated by externally generated radiofrequency signals. Experiments on the magnetic drive and imaging aspects of this system have been conducted in vitro and in vivo. Mechanical studies of cadaver dog brains revealed that a mean force of 0.07 ± 0.03 N was required to move a 3‐mm diameter sphere through brain at a speed of <1 cm/15 s. A cranial phantom with mechanical properties similar to brain was constructed of gelatin and Plexiglas. With the use of a “neck loop” design drive magnet with a maximum magnetic field strength of 0.10 T, a 3 X 3 mm cylindrical neodymium iron boron thermoceptor was smoothly directed through the phantom in two dimensions. Additional experiments were conducted with a larger drive magnet in five anesthetized dogs. Neodymium iron boron and samarium cobalt thermoceptors of various shapes and sizes were placed into the cerebral cortex through a burr hole, then directed with the drive magnet. Fluoroscopy was used to follow the thermoceptor movements. Thermoceptors were moved smoothly through brain parenchyma in three dogs and in the subdural space in two dogs. All dogs tolerated the procedure well; none developed neurological deficits. Histological examination of the brains showed thermoceptor tracts, but no evidence of significant hemorrhage. These results demonstrate the technical feasibility of the magnetic drive aspect of the stereotaxic hyperthermia system. Plans for further research and development are outlined. (Neurosurgery24:444‐448, 1989)