The effects of various depths of barbiturate and halothane anesthesia and different arterial blood carbon dioxide tensions (PaCO2) at uniform levels of anesthesia on brain pH, tissue indicator perfusion, and blood flow were studied in 40 cats. Brain pH was measured using a Iipid-soluble,pH-sensitive fluorescent indicator (umbelliferone), and its clearance was determined from the slope of its washout curve. Cerebral blood flow (CBF) was determined from the clearance of intra-arterially injected xenon-133. Values for CBF were higher in the halothane-exposed group, and with deep anesthesia they did not decrease as much as did those seen with pentobarbital (47 vs. 27 ml/100 g/min). The barbiturate brain pH-arterial blood PaCO2 regression line had a steeper slope than the corresponding halothane line. The two lines crossed at PaCO2 42 torr and brain pH 7.15. Brain pH was directly related to both PaCO2 and depth of anesthesia in the halothane-exposed group but only to PaCO2 in the barbiturate exposed group. If light halothane anesthesia (0.1 per cent) can be considered almost equivalent to the waking state, then both anesthetics produce relative brain alkalinity. The clearance of the pH indicator was only modestly sensitive to changes in PaCO2 at anesthetic levels of pentobarbital and halothane. It was not significantly changed by increasing the amount of pentobarbital. However, increasing levels of halothane produced a decrease of more than 50 per cent in its clearance (63 ml/100 g/min at 0.1 per cent vs. 29 ml/100 g/min at 3 per cent). It is concluded that umbelliferone provides a reliable method for the measurement of brain pH and possibly a useful tool for studies of the blood-brain barrier.