To gain information on the role of central 5-HT2 receptors in the reduction of food intake caused by D-fenfluramine in rats, different intraperitoneal doses of metergoline, a non-selective 5-HT receptor antagonist and ritanserin, a selective 5-HT2 receptor antagonist, were compared for their ability (a) to antagonize the anorectic effect of D-fenfluramine; (b) to occupy central 5-HT2 receptors in vivo (measured by the binding of [3H]spiperone in the frontal cortex) and (c) to affect the concentrations of D-fenfluramine and its active metabolite, D-norfenfluramine in brain. Metergoline dose-dependently reduced the effect of D-fenfluramine (2.5 mg/kg i.p.) on food intake, with complete antagonism at 1 mg/kg, a dose which occupies about 50% of cortical 5-HT2 receptors. Ritanserin, at a dose (0.5 mg/kg) causing 50% occupation of 5-HT2 receptors, had no effect on anorexia induced by D-fenfluramine and only partially prevented it at doses which caused maximum occupation of 5-HT2 receptors (1-2 mg/kg). Unlike 1 mg/kg metergoline, 1 mg/kg ritanserin significantly reduced the concentrations of D-norfenfluramine in the frontal cortex and hypothalamus of rats 30 min after injection of D-fenfluramine. The results suggest that 5-HT receptors, other than 5-HT2, possibly 5-HT1B, are involved in the anorectic effect of D-fenfluramine in food-deprived rats.