The pharmacological activities of haloperidol decanoate in several behavioral and biochemical tests for neuroleptics were pursued to validate its clinical mode of action. Haloperidol decanoate itself was about 40 times less potent than haloperidol in inhibiting 3H-spiperone binding to rat striatum in vitro and its direct application into the cerebral ventricle of mice showed no effect on brain concentrations of dopamine metabolites, homovanillic acid (HVA) and 3,4-dihydroxyphenylacetic acid (DOPAC). When injected intramuscularly, haloperidol decanoate exhibited various behavioral and biochemical effects together with prolonged accumulation of the parent drug haloperidol in dopamine-rich areas of the brain, but there was a clear difference in the time-course patterns of the effects. Conditioned avoidance responses in mice and rats were suppressed for a long period of days in the time-course pattern parallel to the elevation of brain haloperidol levels; the effect reached the peak a couple of days after the treatment and remained there over a week with gradual recovery towards 1 month after the injection. On the other hand, tolerance developed as rapidly as within 7 days to the antagonistic effects on apomorphine-induced stereotyped behaviours in mice and rats and to prolactin-releasing activity in rats. Biochemically, there was a regional difference in the brain in the time-course of effect of haloperidol decanoate. Namely, the initial increase in dopamine metabolites was no longer seen in the striatum on day 21, whereas it still persisted in the frontal cortex. Thus, the present study has confirmed on the one hand the assumption that haloperidol decanoate is a pro-drug which itself has no neuroleptic activity and validated on the other hand the clinical results that it exhibits long-lasting antipsychotic activities with minimum and tolerable extrapyramidal and prolactin-related side effects.