Cerebrovascular autoregulation assures constancy of cerebral perfusion despite blood pressure changes, as long as mean blood pressure remains in a range between 50-170 mmHg. Static and dynamic myogenic mechanisms dampen sudden blood pressure changes. Neurogenic influences of sympathetic, noradrenergic fibers modulate primarily proximal, large diameter segments of cerebral arteries, but also small 15-20 microns diameter vessels. Parasympathetic, vasodilating impulses are of less influence. Monoaminergic brainstem centers such as the dorsal raphe nucleus, locus coeruleus or nucleus reticularis pontis oralis also influence vessel tone. Metabolic, local parenchymal and endothelial substances have major impact on cerebral vessel tone. Particularly important are nitric oxide, calcitonin gene related peptide, substance P, endothelin, potassium channels and autocoids such as histamine, bradykinin, arachidonic acid, prostanoids, leucotrienes, free radicals or serotonin. The clinical examination of autoregulation is mostly based on brief blood pressure changes induced by drugs such as angiotensin, phenylephrine or sodium nitroprusside, or by challenge maneuvers. Frequently, blood pressure is challenged by a tilt-table maneuver, the "leg-cuff"-method according to Aaslid, or a Valsalva maneuver. The analysis of coherence and phase relation between spontaneous or metronomic breathing modulation of blood pressure and brain perfusion also assesses autoregulatory function. Cerebral blood flow is determined by means of transcranial Doppler sonography, mostly of the proximal segment of the mid-cerebral artery. There is some controversy whether a decrease of cerebral blood flow velocity measured at this segment indicates vasodilatation at the insonated segment or reflects blood flow reduction due to decreased perfusion of down-stream vessel segments. Various clinical and animal studies are presented demonstrating diameter constancy of the insonated mid-cerebral artery segment and thus indicating that slowing of mid cerebral artery blood flow velocity as assessed by transcranial Doppler sonography is due to a decrease of down-stream perfusion. Direct, intraoperative measurements of vessel diameter confirm this conclusion.