The anti-immobility effect of the selective melatonin receptor antagonist, luzindole, was investigated in the behavioral despair test using three different strains (C3H/HeN, C57BL/6J and albino ND/4) of mice. The time of immobility of the C3H/HeN during the 240 s swimming period measured at noon (12:00 to 14:00 h) was 47.8 +/- 3.0 s (n = 63) and at midnight (00:00 to 02:00 h) was 67.7 +/- 2.8 s (n = 68) (P less than 0.001, when compared with the noon value), when the levels of endogenous melatonin are presumably low and high, respectively. Melatonin (30 mg/kg) given i.p. did not modify the time of immobility at either time of measurement. Luzindole (30 mg/kg i.p.) reduced the time of immobility in a dose-dependent manner, the effect being more pronounced at midnight (60% reduction) than at noon (39% reduction). The effect of luzindole was time-dependent, showing a maximal effect at 60 min. The anti-immobility effect of luzindole (10 mg/kg i.p.) was prevented by the administration of melatonin (30 mg/kg i.p.). Luzindole (30 mg/kg i.p.) did not modify the time of immobility either at noon or midnight in the albino ND/4 mouse, or in the C57BL/6J mouse, which does not produce melatonin. Our results suggest that endogenous melatonin plays a role during swimming in the C3H/HeN mouse behavioral despair test. We conclude that luzindole may exert antidepressant-like activity in the C3H/HeN mouse by antagonizing the action of endogenous hormone.