A review of the literature suggests that the dorsal hippocampal serotonergic system, and, in particular, the postsynaptic 5-HT(1A) receptor, mediates an anxiogenic response, whereas endogenous dorsal hippocampal cholinergic tone mediates an anxiolytic response. Accordingly, it has been shown that direct dorsal hippocampal administration of the 5-HT(1A) receptor agonist, 8-OH-DPAT, the nicotinic receptor antagonist, mecamylamine, and the M(1) muscarinic receptor antagonist, pirenzepine, all have anxiogenic effects in rats tested in the social interaction test. It is therefore surprising that nicotine also has an anxiogenic effect in this test following dorsal hippocampal administration. However, the anxiogenic effects of mecamylamine and nicotine in the dorsal hippocampus are blocked by coadministration of the 5-HT(1A) receptor antagonist, WAY 100635, suggesting that both of these compounds act by enhancing hippocampal serotonergic transmission, thereby stimulating postsynaptic 5-HT(1A) receptors. This conclusion is supported by the observation that both nicotine and mecamylamine stimulate basal [3H]-5-HT release from dorsal hippocampal slices. A possible mechanism by which nicotinic receptor ligands modulate hippocampal 5-HT release is discussed, and it is proposed that the dorsal hippocampal serotonergic and cholinergic systems are tightly coupled and function antagonistically in the modulation of anxiety, as measured in the social interaction test. These systems are relatively unimportant in controlling behaviour on trial 1 in the plus-maze. On trial 2 in the elevated plus-maze, a model of specific phobia, the endogenous cholinergic system, nicotine, and the M(1) receptor agonist, McN-A-343, all mediate an anxiolytic effect, whereas stimulation of 5-HT(1A) receptors mediates an anxiogenic effect. It is proposed that the hippocampus may predominantly control the avoidance components of phobic anxiety, with other regions, such as the dorsomedial hypothalamus, controlling the escape components.