In accordance with a previous study of adjuvant effects of the anticonvulsant carbamazepine (CBZ) on the neuroleptic treatment of schizophrenic psychoses, the effects of valproate (VPA) were tested in a randomly assigned double-blind, placebo-controlled study. Apart from a (statistically nonsignificant) psychopathological deterioration following discontinuation of VPA while on continuous neuroleptic mediation after four weeks and a statistically significant effect on "hostile belligerence", no overall therapeutic effects of the combination of haloperidol (HPD) with VPA were observed under controlled conditions. Unlike the results with CBZ, concomitant use of VPA led to an even higher consumption of haloperidol and biperiden and to a higher rate of extrapyramidal symptoms compared with the corresponding placebo group, although these differences did not attain statistical significance. In regard to use of the sedative neuroleptic chlorprothixene, there was a trend toward lower doses in the VPA group than in the placebo group. From these results, adjuvant effects like those of carbamazepine in the neuroleptic treatment of schizophrenic psychoses could not be confirmed for valproate in the present study. However, the trend toward lower doses of sedative medication and observed effects on "hostile belligerence" may indicate sedative and/or antimanic properties of valproate which have recently been demonstrated in several controlled studies.