Brain function depends exquisitely on oxygen for energy metabolism. Measurements of brain tissue oxygen tension, by a variety of quantitative and qualitative techniques, going back for >50 years, have led to a number of significant conclusions. These conclusions have important consequences for understanding brain physiology as it is now being explored by techniques such as blood-oxygen-level-dependent functional magnetic resonance imaging (BOLD fMRI) and near-infrared spectroscopy (NIRS). It has been known for some time that most of the measured oxygen tensions are less than venous pO2 and are distributed in a spatially and temporally heterogeneous manner on a microregional scale. Although certain large-scale methods can provide reproducible average brain pO2 measurements, no useful concept of a characteristic oxygen tension or meaningful average value for brain tissue oxygen can be known on a microregional level. Only an oxygen field exists with large local gradients due to local tissue respiration, and the most useful way to express this is with a pO2 distribution curve or histogram. The neurons of the brain cortex normally exist in a low-oxygen environment and on activation are oxygenated by increases in local capillary blood flow that lead to increases in hemoglobin saturation and tissue oxygen.