The literature regarding the blood--brain barrier (BBB) transport of butanol is conflicting as studies report both incomplete and complete extraction of butanol by the brain. In this work the BBB transport of both [14C]butanol and [3H]water was studied using the carotid injection technique in conscious and in ketamine- or pentobarbital-anesthetized rats employing N-isopropyl-p-[125I]iodoamphetamine ([125I]IMP) as the internal reference and as a fluid microsphere. The three isotopes (3H, 125I, 14C) were conveniently counted simultaneously in a liquid scintillation spectrometer. IMP is essentially completely sequestered by the brain for at least 1 min in conscious rats and for 2 min in anesthetized animals. Butanol extraction by rat forebrain is not flow limited but ranges between 77 +/- 1 and 87 +/- 1% for the three conditions. The incomplete extraction of butanol by the forebrain is due to diffusion restriction of butanol clearance in some regions (frontal cortex, colliculi) but not in others (caudate, hippocampus, olfactory bulb). The permeability-surface area product/cerebral blood flow ratio of butanol and water in rat forebrain remains relatively constant, 1.7 +/- 0.2 and 1.0 +/- 0.1, respectively, despite a twofold increase in cerebral blood flow in conscious relative to pentobarbital-anesthetized rats. The absence of an inverse relationship between flow and butanol or water extraction is consistent with capillary recruitment being the principal mechanism underlying changes in cerebral blood flow in anesthesia. The diffusion restriction of BBB transport of butanol in some regions, but not in others, necessitates a careful regional analysis of BBB permeability to butanol prior to usage of this compound as a cerebral blood flow marker.