A new antidepressant, mianserin (GB-94, Organon), and other known antidepressants, imipramine (IMP), desipramine (DPM), chlorimipramine (CIP), amitriptyline (ATP) and nortriptyline (NTP) were orally administered to respective groups of 25 mice for consecutive weeks, and changes in monoamine metabolisms in the brain were investigated, neurochemically. On the basis of significant difference from the control group, mianserin 20 mg/kg and DPM 20 mg/kg produced no change of the brain contents of DA and 5-HT; IMP 50 mg/kg and CIP 40 mg/kg did not alter NA and DA values, but did decrease 5-HT levels; ATP 30 mg/kg increased only the amount of NA, left unchanged the DA and 5-HT levels, and NTP 20 mg/kg raised the levels of DA and 5-HT, but had no effect on NA. Mianserin, IMP and DPM all facilitated the decrease of both NE and DA as induced by pre-administration of alpha-methyltyrosine, the substance known to reduce the level of catecholamines. Both NTP and CTP showed no notable effect. ATP suppressed the reduction of the DA level. Accumulation of 5-HIAA induced 1 hour after treatment with probenecid 250 mg/kg was enhanced by mianserin 20 mg/kg, NTP 20/kg and IMP 25 mg/kg, but was not significantly affected by CIP, DPM and ATP. These results suggest that mianserin activated both the NE and 5-HT metabolic systems, IMP activated the NE system in general, but activated the 5-HT system only at low doses and tended to suppress the formation of 5-HT, in high doses; CIP had no effect on the NE system and suppressed 5-HT system, in high doses. The arousal response of hippocampal EEG in the chronic EEG of the rabbit, induced by physostigmine 0.05 mg/kg, i.v., was suppressed by IMP, ATP and CIP, while DMP and NTP were uneffected. Mianserin, rather maintained and prolong the arousal reaction. Observations made on arousal and sleep phases in long time recording of chronic spontaneous EEG in the rat showed a dose-dependent shortening of the para-sleep phase by mianserin in doses over 30 mg/kg and IMP in doses over 5 mg/kg. ATP 15 mg/kg showed a tendency to shorten the para-sleep. Hippocampal and cerebral cortical EEGs at arousal and slow wave stages were analyzed using a signal processor. The power spectra thus obtained showed no significant difference between the effects of mianserin and IMP.