The unique place of the RCT in the evaluation of medical therapeutics is well documented and accepted by those dedicated to the scientific advancement of health care. It is best accepted in the evaluation of new drugs, in part, because there is a large commercial community with sufficient resources to support such trials (and a substantial economic incentive to do so) and, in part, because there is strong regulation of the pharmaceutical industry requiring such information before a drug can be marketed. Surgery, in general, and neurosurgery, in particular, have been slow to adopt the techniques of biostatistics, including the RCT. The typical analytic techniques used in neurosurgery (Student's t test, χ2 test, ANOVA) were developed contemporaneously with neurosurgical techniques such as sphygmomanometry, electrocautery, and ventriculography.16 The RCT was introduced into medical research at a time when neurosurgeons were just learning to use myelography. Both methods of validation have advanced remarkably since those times, but at the beginning of the third millenium, we evaluate our practice with 1920s techniques. The scientific rationale for RCTs is well established. The feasibility of such trials of surgical procedures is demonstrated. The ethical imperative to evaluate a dangerous intervention before submitting large numbers of patients to it is clear. With increasing economic pressure on the health care system, the external demand (expressed through economic controls on surgical intervention) for better justification for operations is likely to lead to more frequent application of the RCT in neurosurgery.