Selective brain cooling (SBC) is defined as a brain temperature cooler than the temperature of arterial blood from the trunk. Surrogate measures of arterial blood temperature have been used in many published studies on SBC. The use of a surrogate for arterial blood temperature has the potential to confound proper identification of SBC. We have measured brain, carotid blood, and rectal temperatures in conscious sheep exposed to 40, 22, and 5 degrees C. Rectal temperature was consistently higher than arterial blood temperature. Brain temperature was consistently cooler than rectal temperature during all exposures. Brain temperature only fell below carotid blood temperature during the final few hours of 40 degrees C exposure and not at all during the 5 degrees C exposure. Consequently, using rectal temperature as a surrogate for arterial blood temperature does not provide a reliable indication of the status of the SBC effector. We also show that rapid suppression of SBC can result if the animals are disturbed.