Background Cortical acetylcholine originates in the basal forebrain and is essential for maintaining normal cognition and arousal. Morphine impairs these cholinergically mediated cortical functions. The current study tested the hypothesis that morphine decreases prefrontal cortical acetylcholine release by acting at the level of the basal forebrain. Methods Adult male Sprague-Dawley rats (n = 18) were anesthetized with isoflurane. One microdialysis probe was placed in the substantia innominata region of the basal forebrain and perfused with Ringer's solution (control) followed by one concentration of morphine (1, 10, 100, or 1,000 microm) or morphine (1,000 microm) plus naloxone (100 microm). A second microdialysis probe was placed in the prefrontal cortex for measuring acetylcholine. In a second series of experiments, rats (n = 6) were implanted with electrodes for recording states of arousal, a guide cannula positioned above the prefrontal cortex for inserting a microdialysis probe, and an indwelling jugular vein catheter. The effects of administering intravenous morphine (30 mg/kg) versus normal saline (0.9%) on prefrontal cortical acetylcholine release, cortical electroencephalographic power, and behavior were quantified. Results Dialysis delivery of morphine to the substantia innominata caused a concentration-dependent, naloxone-sensitive decrease in acetylcholine release within the prefrontal cortex. The maximal decrease in acetylcholine was 36.3 +/- 11.5%. Intravenous morphine administration significantly decreased cortical acetylcholine release, increased electroencephalographic power in the 0.5- to 5-Hz range, and eliminated normal wakefulness. Conclusion Morphine causes obtundation of arousal and may cause cognitive impairment by acting at the level of the substantia innominata to disrupt cortical cholinergic neurotransmission.