We know that general anesthesia produces unconsciousness but not quite how. We recorded neural activity from the frontal, parietal, and temporal cortices and thalamus while maintaining unconsciousness in non-human primates (NHPs) with propofol. Unconsciousness was marked by slow frequency (~1 Hz) oscillations in local field potentials, entraining local spiking to Up states alternating with Down states of little spiking, and decreased higher frequency (>4 Hz) coherence. The thalamus contributed to cortical rhythms. Its stimulation "awakened" anesthetized NHPs and reversed the electrophysiologic features of unconsciousness. Unconsciousness thus resulted from slow frequency hypersynchrony and loss of high-frequency dynamics, partly mediated by the thalamus, that disrupts cortical communication/integration. Unconsciousness is general anesthesia's most intriguing feature 1 . In humans, unconsciousness by propofol-a widely used anesthetic-is linked to structured neurophysiological dynamics: decreases in higher frequency (>15 Hz) but increases in frontal alpha (8 to 12 Hz) 2-5 and slow frequency (0.1 to 1 Hz) oscillations that phaselock neuronal spiking 6 and loss of fronto-parietal connectivity. 7 Nevertheless, the details remain unknown. These dynamics have been investigated using extracranial measures (EEG, fMRI) with limited spatial or temporal specificity or microelectrode recordings with limited spatial coverage in patients. They have yet to be studied at high spatial and temporal resolutions simultaneously. Moreover, previous studies have not included the thalamus, a critical nexus that regulates cortical activity 8,9 .A vascular access port implanted subcutaneously in two non-human primates (NHPs) allowed us to infuse propofol to transition them from awake through loss of consciousness (LOC) and recovery of consciousness (ROC) without interruption of neurophysiological recording (and without introduction of other anesthetics or other confounding factors, see Methods, Figure 1A). We started with a high dose for 15 minutes (0.28-0.58 mg/kg/min, adjusted per individual animal). After LOC (indicated by cessation of airpuff-evoked eyeblinks, reduction in heart rate, pupil dilation, and decreased muscle tone), infusion was lowered to a holding dose (0.14-0.23 mg/kg/min). After infusion ceased, ROC followed after ~60 sec. We recorded spikes and local field potentials (LFPs) from 64-channel "Utah" arrays in frontal cortex (8A, PFC), posterior parietal (PPC, 7A/B), and auditory/temporal (Superior Temporal Gyrus, STG) cortex and LFPs from multiple-contact stimulating electrodes (two in each hemisphere) in frontal thalamic nuclei (intralaminar nuclei, ILN, and mediodorsal, MD, Figure 1B).We compared LFP power during the pre-drug Awake state (-120 to 0 seconds pre drugonset) to every timepoint after drug administration (non-parametric cluster-based randomizations, corrected for multiple comparisons; all effects p<0.01 10 ). Around 60-75 seconds post-drug administration (pre-LOC), theta/alpha (4-15 Hz) power decreased in Fi...