Rats were raised from weaning (21 days old) to young adulthood (50-60 days old) in either an enriched or impoverished stimulus environment. In the enriched condition (EC), rats were group-housed with various novel objects that were re-arranged daily. In the impoverished condition (IC), rats were housed individually without any objects. As adults, a four-trial conditioned place preference (CPP) test was used to assess locomotor activity and reward produced by morphine (0, 0.1, 1 or 10 mg/kg). On morphine conditioning day 1, both EC and IC rats displayed an inverted U-shaped dose-effect curve for locomotor activity and the locomotor stimulant effect of acute morphine was greater in IC than EC rats. Across morphine conditioning days 1-4, both EC and IC rats displayed locomotor sensitization; the locomotor sensitization following repeated morphine injections was greater in IC than EC rats. In contrast to the enhanced locomotor stimulant effect of morphine observed in IC rats, morphine-induced CPP was attenuated in IC rats relative to EC rats, indicating that the locomotor and rewarding effects of opioids depend upon different neural substrates. Measurement of mu opioid receptor density and rates of morphine-stimulated dopamine synthesis in the mesolimbic and nigrostriatal systems of EC and IC rats revealed no reliable differences between groups. Therefore, the ability of mu opioid receptors to modulate mesolimbic dopamine neurotransmission does not account for the differential behavioral effects of morphine in EC and IC rats.