Galantamine, currently used in Alzheimer's patients, has shown neuroprotection in hippocampal slices subjected to oxygenglucose deprivation. Here, we present an in vivo study to evaluate the potential neuroprotective effects of galantamine in a transient global cerebral ischemia model in gerbils. Three treatment protocols were used. In the pretreatment protocol, gerbils were treated before ischemia and for 3 consecutive days thereafter. Eight groups of animals were included: sham operation plus placebo, 10 mg/kg mecamylamine and 10 mg/kg galantamine, respectively; and ischemia plus placebo, 10 mg/kg mecamylamine, 1 mg/kg galantamine, and 10 mg/kg galantamine and 10 mg/kg mecamylamine plus galantamine, respectively. Postischemia protocols included three groups of animals: sham operation, ischemia plus placebo, and ischemia plus 10 mg/kg galantamine; substances were administered 3 or 6 h after ischemia and for 2 consecutive days thereafter. Pyramidal neurons surviving in the cornus ammonis 1 region of the hippocampus were evaluated 72 h after reperfusion, terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) histochemistry, caspase-3 and superoxide dismutase (SOD)-2 immunohistochemistries, and Western blottings were performed, and object placement tests were carried out. Galantamine significantly increased the number of living pyramidal neurons after ischemia-reperfusion injury. Galantamine significantly reduced TUNEL, active caspase-3, and SOD-2 immunoreactivity. The nicotinic antagonist mecamylamine blocked the protective effects of galantamine. The neuroprotective effects of galantamine were preserved even when first administered at 3 h postischemia. These results correlated with the performance in the object placement test. This study shows that galantamine provides in vivo neuroprotection and memory recovery against global cerebral ischemia, even when administration begins 3 h postischemia.