Long-Evans female rats were "trained" in an 8-arm radial maze and subsequently tested under systemic treatment with physostigmine (0.05 mg/kg, IP), scopolamine methylbromide (MBr) and scopolamine hydrobromide (HBr; 0.5 mg/kg, IP), whose effects were compared to those of aspirative lesions of the fimbria-fornix pathways. During the predrug trials, rats with lesions showed impaired performances compared to those of intact rats. Whereas physostigmine had no significant effect in either group, scopolamine HBr impaired performances of intact rats in a manner closely parallel to all measured behavioral effects of the lesions (errors, "correct arms" and strategies). The scopolamine HBr-induced deficits were not correlated with the percentage of "spatial" strategies. Under scopolamine HBr treatment the performances of rats showing preferences for "spatial" strategies did not differ significantly from those of rats showing preferences for "orientation" strategies. These results provide further support for the involvement of cholinergic processes in working memory and suggest that scopolamine-induced central cholinergic disruption may mimic the effects of fimbria-fornix lesions in an 8-arm radial maze. They also somewhat qualify previous reports on 1) the poor sensitivity of an uninterrupted radial maze testing procedure to pharmacological treatment and 2) the abilities of rats to resist muscarinic blockade depending on the strategies they use in the maze.