The conditions under which a drug is administered often alter the behavioral effects of that drug. The present study examined the effect of changes in response dependence on the discriminative stimulus effects of ethanol. Six Long-Evans rats were trained to discriminate 1000 mg/kg, interperitoncal (i.p.) ethanol from saline. A dose—effect curve was then obtained using i.p. doses of 100, 320, 560, 1000, 1320 and 1560mg/kg ethanol. Ethanol doses of 1000mg/kg and greater produced more than 80% ethanol-lever selection. The rats were then trained to orally self-administer 10% weight/volume ethanol and tested to determine if self-administered oral ethanol would substitute for experimenter administered i.p. ethanol. A mean self-administered ethanol intake of 1114mg/kg (±156mg/kg) produced 83% ethanol-lever responding. Restricted access to 560 mg/kg of self-administered ethanol resulted in 33% i.p. ethanol-lever responding. Doses of 100 and 320mg/kg ethanol did not substitute for i.p. ethanol. These data show that orally self-administered ethanol can produce discriminative stimulus effects that are similar to i.p. experimenter-administered ethanol and that orally self-administered ethanol produces centrally-mediated discriminative stimulus effects. Behav Pharmacol 1998; 9:329–336 © 1998 Lippincott-Raven Publishers.