The diagnostic accuracy of CT in cases of lumbar disk prolapse was investigated on the basis of a group of 158 of our own patients who were divided into three separate groups. CT findings were compared on the one hand with myelograms and on the other with results of surgery. Simultaneously performed myelographic and CT examinations produced different diagnostic findings in 7 out of 80 lumbar disks (approx. 9%). These deviating findings ranged from gradual differences (in 3 cases) to completely contradictory findings (in 4 cases). In 31 patients who underwent surgery, approximately the same amount of false-positive and false-negative findings were obtained in preoperative diagnosis using CT as with myelography. The overall accuracy calculated for non-invasive computerized tomography was 93.7%, while for myelography it was 90.7%. Out of a further 91 patients in whom myelography was not performed and whose condition was checked surgically, the diagnosis had been established correctly by CT in 88. This represents a degree of accuracy of 96%. The vertical and lateral location of the disk prolapses as determined by CT corresponded with findings at surgery in every case. It is concluded from the results of these investigations that CT should be given absolute priority over myelography in the diagnosis of lumbar disk prolapse and that if there is a correlation between clinical history, neurological deficiency symptoms, and CT findings, myelography is unnecessary, even for surgery. Only if there are discrepancies between the findings myelography is indicated as a supplementary examination. CT and myelography are examination methods which complement, rather than complete with one another.