THE IMPORTANCE of collateral circulation in the brain has been evident since the observations of Astley Cooper on ligation of the common carotid artery in man 1 and ligation of the principal cerebral vessels in the dog.2It is also common neurological experience that there is considerable variation in capacity for adjustments in the cerebral circulation from one person to another, for carotid occlusion in one person may occur without symptoms, whereas the infarction of most of a cerebral hemisphere may follow in another. The anatomical explanation for this is, in part, the variation in size of the communicating arteries of the circle of Willis. Such variation of collateral circulation prompts the neurosurgeon to compress the carotid artery preoperatively as a test of collateral circulation before ligating this vessel. Recent experience with arteriography has shown that some persons may successfully survive for years without either one or the other carotid artery functioning, rarely without both. In later years episodes of cerebral ischemia occur as the collateral circulation becomes restricted by further vascular disease or by falls in the systemic blood pressure.3 Records (in man) of intravascular pressure at the distal end of the carotid artery, in the circle of Willis, and in the middle cerebral vessels seldom show a fall greater than 30 to 50% when the carotid artery is occluded.* Varying degrees of recovery in the first few days following occlusion of the smaller cerebral vessels are also seen frequently. Lack of compensation in some cases raises the question as to how much adjustment may be due to smaller anasto¬ moses on the surface of the brain, the importance of which has recently been emphasized,6 though their presence was known to the earlier anatomists.In order to obtain some information concerning the mechanisms underlying this compensation in collateral cerebral vessels and their site of action, we have sought a method of plotting the effectiveness of circulation in several areas of cortex at the same time. We have chosen the polarographic method of measuring local changes in tissue oxygénation (EPG), using multiple platinum electrodes, along the lines of the investigation of coronary occlusion by Sayen and co-workers.7