The anticonvulsant carbamazepine (100 mg/kg i.p.) increased the levels of 3-methoxy-4-hydroxyphenylglycol sulfate in the rat brain as well as its accumulation after inhibition of acid transport by probenecid. It did not interfere with 3-methoxy-4-hydroxyphenylglycol sulfation in vivo. However, this was not corroborated by the results of turnover studies. The depletion of brain norepinephrine after inhibition of tyrosine hydroxylase or dopamine (DA) beta-hydroxylase rather tended to be decreased and the accumulation of DA in the rat heart after DA beta-hydroxylase inhibition was diminished. Carbamazepine also reduced the depletion of striatal DA after tyrosine hydroxylase inhibition and the accumulation of striatal dopa after central decarboxylase inhibition, but did not affect the levels of the deaminated DA metabolites, homovanillic acid and 3,4-dihydroxyphenylacetic acid, even after pretreatment with probenecid. Therefore, carbamazepine seems to reduce catecholamine turnover in spite of its enhancing action on the firing of norepinephrine and DA neurons reported in the literature. These apparently paradoxical effects might be explained by an interference with the catecholamine storage mechanism. Two other anticonvulsants, diphenylhydantoin and phenobarbital, in doses approximately equieffective to that of carbamazepine with respect to their anticonvulsant action, showed similar although somewhat weaker effects on DA, but not on norepinephrine turnover. Within the catecholamine hypothesis of affective disorders, the reduction of catecholamine turnover by carbamazepine might explain the reported antimanic and antipsychotic effects in patients.