Ischemic liver tissue was produced by clamping the portal venous and hepatic arterial blood supply to the left lateral and median lobes of rat liver. If, after 2 to 3 hours of ischemia, reflow to the liver was established by removing the clamp, two-thirds or more of the liver cells were histologically dead 24 hours later. Pretreatment with chlorpromazine (20 mg/kg) 30 minutes before inducing ischemia for up to 3 hours virtually completely prevented this ischemic cell death. If the animals were kept alive for an additional 24 hours with no further treatment, the extent of liver cell necrosis at 48 hours was still markedly less than that seen in the untreated ischemic controls. Administration of chlorpromazine after induction of ischemia and immediately prior to the onset of reflow reduced but did not completely prevent ischemic cell death as determined at 24 hours. This protective action of chlorpromazine was confirmed by the ability of the treated animals to regenerate cellular ATP levels after 3 hours of ischemia. In addition, chlorpromazine was shown to significantly reduce the increases in total liver cell and mitochondrial calcium ion contents that accompany the return of blood flow to irreversibly injured liver cells. The protective effect of chlorpromazine could not be attributed to any effect either on the rate or extent to which the liver cells became ischemic or on the perfusion patterns following release of the obstruction, and it is concluded that the action of chlorpromazine must be on some component(s) of the reaction of the cells to the ischemia itself. The possible basis of this action is discussed.