1. SB 200646A, N-(1-methyl-5-indolyl)-N'-(3-pyridyl) urea hydrochloride, the first reported selective 5-HT2C/2B over 5-HT2A receptor antagonist, (pK1 rat 5-HT2C receptor 6.9, pA2 rat 5-HT2B receptor 7.5, pK1 rat 5-HT2A receptor 5.2) dose-dependently blocked a putative rat model of 5-HT2C receptor activation; 1-(3-chlorophenyl)piperazine (mCPP, 5 mg kg-1, i.p. 20 min pretest)-induced hypolocomotion (estimated ID50 19.2 mg kg-1, p.o.). 2. SB 200646A also blocked another putative in vivo model of 5-HT2C receptor function; mCPP (5 mg kg-1, i.p. 20 min pretest)-induced hypophagia in 23 h food-deprived rats (estimated ID50 18.3 mg kg-1, p.o.). 3. SB 200646A did not antagonize 1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane (DOI)-induced head shakes in rats at doses up to 200 mg kg-1, p.o., an effect thought to be mediated by 5-HT2A receptors for which SB 200646A has its next highest affinity (50 fold less) after the 5-HT2C and 5-HT2B sites. 4. SB 200646A (20, 40 mg kg-1, p.o., 1 h pretest) also reversed mCPP (0.5 mg kg-1, i.p., 30 min pretest)-induced anxiety in the social interaction test, under low light familiar conditions. 5. When given alone, under high light unfamiliar conditions, SB 200646A (2-40 mg kg-1, p.o.) increased active social interaction without affecting locomotor activity in the rat social interaction test. This is consistent with an anxiolytic action of SB 200646A. 6. These results indicate that SB 200646A has in vivo efficacy and that 5-HT2C or 5-HT2B receptors are indeed likely to mediate mCPP-induced hypolocomotion, hypophagia and anxiogenesis. They also suggest that 5-HT2C,2B receptor blockade induces anxiolysis.