In the present experiments, we investigated the influence of chronic modifications of circulating levels of corticosterone on the locomotor response to amphetamine. Different groups of rats were adrenalectomized and implanted subcutaneously with pellets releasing different amounts of corticosterone (0-200 mg). A wide range of corticosterone concentrations was reached in order to saturate selectively either the type I (mineralocorticoid) or the type II (glucocorticoid) corticosteroid receptors. The locomotor response to d-amphetamine (1.5 mg/kg) was studied 10-14 days later. We found that adrenalectomy reduced the response to d-amphetamine by 33% and that a normal response was restored with pellets releasing physiological concentrations of corticosterone (50-mg pellets), and was potentiated in animals with pellets releasing high amounts of corticosterone mimicking chronic stress situations (200-mg pellets). The correlation between plasma corticosterone concentration, locomotor activity following d-amphetamine and thymus weight, which is a reliable indicator of glucocorticoid action, shows that the influence of the locomotor response to d-amphetamine administration is likely to be mediated via a type II receptor. Since the locomotor activating effect of peripheral administration of d-amphetamine has been shown to depend on the integrity of the dopaminergic innervation of the nucleus accumbens, the effect of d-amphetamine at different doses (0, 1, 3, 10 micrograms/microliter) injected directly into the nucleus accumbens was studied. The results demonstrated that removing the circulating corticosterone induced a similar decrease of the locomotor activity elicited by d-amphetamine injection in the nucleus accumbens. This response was restored in animals with the 50- and 200-mg pellets.(ABSTRACT TRUNCATED AT 250 WORDS)