Abstract The nicotinic antagonist mecamylamine has been found to impair working memory performance in the radial-arm maze (RAM) after s.c. or i.c.v. administration. Mecamylamine has important interactions with dopaminergic (DA) systems. Mecamylamine-induced memory deficits in the RAM are potentiated by the D 2 antagonist raclopride and reversed by the D 2 agonist quinpirole. The nicotinic agonist nicotine has been found to improve working memory performance in the RAM after s.c. or i.c.v. administration. Nicotine-induced memory improvement in the RAM is potentiated by the D 2 agonist quinpirole. The midbrain DA nuclei, the substantia nigra (SN) and the ventral tegmental area (VTA) have relatively dense concentrations of nicotinic receptors which may be critical sites of action for mecamylamine and nicotine. In the current study, the effects of mecamylamine (1, 3.3 and 10 μg/side) infusions into the SN ( n = 12) and VTA ( n = 13) on working memory in the radial-arm maze were examined in adult female Sprague-Dawley rats. The 10-μg/side dose of mecamylamine significantly impaired radial-arm maze working memory performance when infused into either the SN or VTA. No significant effects of mecamylamine on response latency were seen. The nicotinic agonists cytisine (0.1, 0.33 and 1.0 μg/side) and nicotine (0.3, 1.0 and 3.3 μg/side) were administered in a counterbalanced order. The high dose of cytisine (1 μg/side) nearly caused a significant deficit in choice accuracy. Nicotine slightly depressed choice accuracy but not significantly in this study. The interaction of nicotine and mecamylamine was then studied. A dose of 1.0 μg/side of nicotine caused a significant decrease in choice accuracy. Interestingly, this was significantly reversed by a 3.3-μg/side dose of mecamylamine. Studies of the muscarinic antagonist scopolamine (1, 3.3 and 10 μg/side) and the muscarinic agonist pilocarpine (3, 10 and 30 μg/side) did not detect significant effects on RAM choice accuracy. These data support the involvement of nicotinic innervation of the midbrain DA nuclei in memory function.