The effect of the development of tolerance to amphetamine-anorexia on both amphetamine-induced and haloperidol-induced motor effects was investigated. The animals in experiment 1 showed an acute anorexic reaction to 3 mg/kg amphetamine, whereas the rats in experiment 2 failed to meet the criterion level of acute anorexia. During initial training rats received 13 injections of saline or 3 mg/kg amphetamine intraperitoneally (i.p.) every other day. In both experiments, for one group each amphetamine injection was followed 20 min later by 30 min access to milk (CONT groups). In a second group, each amphetamine injection was followed 24 h later by 30 min access to milk (NONCON groups) and a third group received only saline injections and milk (controls). As expected, in experiment 1 originally anorexic animals in the CONT group developed tolerance, whereas the NONCON and control groups displayed no tolerance to amphetamine-anorexia. The NONCON group showed sig nificantly more stereotypy than either the CONT or control group. Furthermore, following 1.25 mg/kg haloperidol the CONT animals were less cataleptic than the NONCON and control groups which did not differ. In experiment 2, at the end of training the rats in all groups displayed no anorexia following amphetamine injection; they consumed an amount of milk equivalent to that normally consumed under no-drug conditions. Neither was there a difference in the amount of catalepsy between groups following injection of 1.25 mg/kg haloperidol.