OBJECTIVE Reports in clinical literature about persistent terminal extrapyramidal hyperkinesis in neuroleptic long-term treated patients and speculation about demential brain decomposition in such cases give rise to support irreversible psychopharmacotoxical brain damage. Histopathological and animal experimental results in this question given up to this day are not in agreement with each other. Therefore an own chlorpromazine long-term experiment in rats is reported with special regard to biometric-statistical results. METHODS Experiments were performed in 36 adult albino rats of either sex, weighing 300-400 g, fed by a standard diet, receiving food and water ad libitum. The animals were divided into 3 groups of 12 animals accidentally. Group I received 15 mg/kg/die chlorpromazine per os by means of a throat probe, group II 10 mg/kg/die, group III was left untreated (controls). The experiment was carried out for 6 months, the animals were killed by perfusion of Bouin's solution 6-8 weeks after interruption of chlorpromazine application. 8 mum paraffin sections were stained by kresylviolet. After histological examination the material was investigated statistically. In the ncl. N. XII, ncl. N. VII, ncl, orig et term. N.V., ncl. cochlearis ant., ncl. vestibularis princeps, Oliva sup., and ncl. dentatus the glia/neurons ratio was stated by counting 200 cells, and in the formation reticularis 400 cells in each animal. The calculated differences in the glia/neurons ratio between the groups were evaluated for statistical significance by the chi2-test. In the ncl. dentatus, oliva sup., ncl, cochlearis ventr., and ncl, vestibularis glia cells and neurons were counted in a plane of 1.2 mm2, in the formatio reticularis and in homologuous parts of cerebellar stratum moleculare in a plane of 2.4 mm2. The differences between the mean values of the groups were verified for statistical significance by means of the t-test. RESULTS In histological examination only 50% of the animals of group I showed a slight loss of neurons and increase of glia cells. Statistically however, significant increase of glia cells was found in the glia/neurons ratio all over the investigated area (rhombencephalic brain stem and cerebellum) and for the ncl. dentatus, the oliva sup., and the ncl. cochlearis ventr. especially (table 1, fig.3). This change in the glia/neurons ratio is caused by a tendency for decreasing of neurons and increasing of glia cells, too (table 2). In the nucleus dentatus the loss of neurons was found to be of high significance. These changes are supposed to be due to chlorpromazine action, and in this manner the experimental results speak for an irreversible psychopharmacotoxical brain damage.