Thalamic, pallidal, and hypothalamic targets were determined in 16 patients by a stereotactic computerized tomography (CT) study using a noninvasive technique with Laitinen's Stereoadapter. At surgery, the Stereoadapter was remounted to the head and the stereotactic CT coordinates were transferred to the Stereoguide without radiography. Air ventriculography was then carried out. The positions of the anterior and posterior commissures (AC and PC), the length of the intercommissural (IC) line, the width of the third ventricle, and the stereotactic coordinates of the target were measured on the ventriculograms and compared to the stereotactic CT measurements. The study showed that the width of the third ventricle was significantly larger on the ventriculograms than on the stereotactic CT scans, whereas the length of the IC line was not significantly different. The differences in the coordinates of the target and of the AC and PC were statistically significant only for the anteroposterior (y) coordinate. Both commissures as well as the surgical target lay, on average, 1.0 mm more anteriorly on the ventriculograms than on the stereotactic CT study. It is concluded that air ventriculography may cause slight anterior displacement of the midbrain structures. The surgical coordinates of the targets based on the stereotactic CT study with the Stereoadapter were on average as accurate as those obtained with ventriculography; therefore, ventriculography may become superfluous in functional stereotaxis.