Serotonin (5-HT) can either increase or decrease anxiety-like behaviour in animals, actions that depend upon neuroanatomical site of action and 5-HT receptor subtype. Although systemic studies with 5-HT(2) receptor agonists and antagonists suggest a facilitatory role for this receptor subtype in anxiety, somewhat inconsistent results have been obtained when such compounds have been directly applied to limbic targets such as the hippocampus and amygdala. The present study investigated the effects of the 5-HT(2B/2C) receptor agonist mCPP bilaterally microinjected into the dorsal hippocampus (DH: 0, 0.3, 1.0 or 3.0nmol/0.2microl), the ventral hippocampus (VH: 0, 0.3, 1.0 or 3.0nmol/0.2microl) or the amygdaloid complex (0, 0.15, 0.5, 1.0 or 3.0nmol/0.1microl) in mice exposed to the elevated plus-maze (EPM). Test sessions were videotaped and subsequently scored for conventional indices of anxiety (percentage of open arm entries and percentage of open arm time) and locomotor activity (closed arm entries). Results showed that mCPP microinfusions into the DH or VH failed to affect any behavioural measure in the EPM. However, when injected into the amygdaloid complex, the dose of 1.0nmol of this 5HT(2B/2C) receptor agonist increased behavioural indices of anxiety without significantly altering general activity levels. This anxiogenic-like effect of mCPP was selectively and completely blocked by local injection of a behaviourally-inactive dose of SDZ SER-082 (10nmol/0.1microl), a preferential 5-HT(2C) receptor antagonist. These data suggest that 5HT(2C) receptors located within the amygdaloid complex (but not the dorsal or ventral hippocampus) play a facilitatory role in plus-maze anxiety in mice.