Convergent evidence suggests that amphetamine (AMPH) exerts its rewarding and locomotor stimulating effects via release of dopamine in the nucleus accumbens. However, there is no consensus as to the relative contributions of core and medial shell subregions to these effects. Moreover, the literature is based primarily on intracranial administration, which cannot fully mimic the drug distribution achieved by systemic administration. In the present study, the effects of bilateral 6-hydroxydopamine lesions of the accumbens core or medial shell on rewarding and locomotor stimulating effects of systemically administered amphetamine (0.75 mg/kg, i.p.) were examined in a conditioned place preference (CPP) procedure relying solely on tactile cues (floor texture). Residual dopamine innervation was quantified by [125I]-RTI-55 binding to the dopamine transporter. When lesions were performed before the conditioning phase, AMPH-induced locomotor stimulation and CPP magnitude were positively correlated with residual dopamine transporter binding in core and medial shell, respectively. Medial shell lesions did not affect morphine CPP, arguing that a sensory or mnemonic deficit was not responsible for the lesion-induced reduction in AMPH CPP. Medial shell lesions performed between the conditioning phase and the test day reduced the expression of amphetamine CPP. These results suggest that after systemic amphetamine administration, rewarding and locomotor stimulating effects of the drug are anatomically dissociated within the nucleus accumbens: the medial shell contributes to rewarding effects, whereas the core contributes to behavioral activation.