1. Most of the intracellular electrophysiological data on cortical neurons has been obtained in anesthetized or reduced preparations, and differs from observations in awake, intact animals. To determine whether these differences are due to experimental techniques or physiological factors, we recorded membrane potentials intracellularly from motor cortex neurons in chronically prepared cats and monkeys under Nembutal-anesthetized, Halothane-anesthetized, and unanesthetized conditions, or during transitions between anesthetized and awake conditions. 2. Resting membrane potentials were found to depend on the anesthetic state of the animal. Membrane potentials of neurons recorded in awake animals were more depolarized than those recorded in the anesthetized state. In the awake state membrane potentials were all less than -65 mV. 3. The input resistance of neurons recorded in awake animals were significantly smaller than those measured in the anesthetized state. Action potentials recorded in awake animals typically showed an undershoot (i.e. negative values at peak), implying that voltage-dependent conductances may be altered. Undershoot of the action potential was more prominent in pyramidal tract neurons (PTNs) than non-PTNs. 4. These data suggested that in awake animals motor cortex neurons, especially PTNs, receive sustained excitatory synaptic input or neuro-modulatory activities.