The reinforcing efficacy of cocaine is thought to involve, at least in part, mesocortical dopaminergic (DA) neurons. Rats will self-administer cocaine applied directly into the medial prefrontal cortex but not into nucleus accumbens or the ventral tegmental area (Goeders & Smith, 1983). The present experiments were conducted to assess whether lesions of prefrontal cortex (mesocortical DA target regions) attenuate the reinforcing properties of systemically administered cocaine. Male Sprague-Dawley rats were anesthetized, and one of three subfields (medial, orbital, or precentral) of the prefrontal cortex was removed by aspiration or no brain injury was done (sham operates). In four experiments the rats were tested on conditioned place preference (CPP), conditioned taste aversion (saccharin conditioned stimulus, cocaine unconditioned stimulus), general activity in the running wheel and open field, and food-reinforced spatial alternation in the T-maze. Sham operates demonstrated a cocaine-induced place preference, rats with medial frontal lesions showed a cocaine-induced place aversion, and other operates showed neither a conditioned place preference nor an aversion. The results of this experiment suggest that lesions of the DA projection fields of the prefrontal cortex in the rat reduce the positive reinforcing properties of systemically injected cocaine. In the second experiment, all subjects showed a conditioned taste aversion of equal magnitude. This suggests that whereas the positive reinforcing properties were affected differentially by prefrontal cortex lesions, the aversive properties were not affected. In Experiment 3 there were no lesion-induced differences in activity in either the running wheel or the open field. Therefore, changes in motor activity cannot account for the CPP data. In the final experiment, the medial frontal operates were impaired relative to the precentral and sham operates on learning to alternate choices in the T-maze, but the orbital frontal operates' performance was not different from that of any other group. This suggests that a general disruption of all reinforcement mechanisms did not occur following these lesions. Instead, these results indicate that mesocortical DA projection regions are involved with mediating the reinforcing properties of cocaine and that there is a separate system mediating the aversive properties of cocaine.