Chronic mild stress (CMS), a well-validated model of depression, was used to study the effects of the melatonin agonist and selective 5-HT(2C) antagonist agomelatine (S 20098) in comparison with melatonin, imipramine, and fluoxetine. All drugs were administered either 2 h before (evening treatment) or 2 h after (morning treatment) the dark phase of the 12-h light/dark cycle. Chronic (5 weeks) evening treatment with agomelatine or melatonin (both at 10 and 50 mg/kg i.p.) dose-dependently reversed the CMS-induced reduction in sucrose consumption. The magnitude and time course of the action of both drugs was comparable to that of imipramine and fluoxetine (both at 10 mg/kg i.p.); however, melatonin was less active than agomelatine at this dose. The effect of evening administration of agomelatine and melatonin was completely inhibited by an acute injection of the MT(1)/MT(2) antagonist, S 22153 (20 mg/kg i.p.), while the antagonist had no effect in animals receiving fluoxetine or imipramine. When the drugs were administered in the morning, agomelatine caused effects similar to those observed after evening treatment (with onset of action faster than imipramine) but melatonin was ineffective. Moreover, melatonin antagonist, S 22153, did not modify the intakes in stressed animals receiving morning administration of agomelatine and in any other control and stressed groups tested in this study. These data demonstrate antidepressant-like activity of agomelatine in the rat CMS model of depression, which was independent of the time of drug administration. The efficacy of agomelatine is comparable to that of imipramine and fluoxetine, but greater than that of melatonin, which had no antidepressant-like activity after morning administration. While the evening efficacy of agomelatine can be related to its melatonin receptors agonistic properties, its morning activity, which was not inhibited by a melatonin antagonist, indicates that these receptors are certainly required, but not sufficient to sustain the agomelatine efficacy. It is therefore suggested that the antidepressant-like activity of agomelatine depends on some combination of its melatonin agonist and 5-HT(2C) antagonist properties.