Buspirone, benzodiazepines, barbiturates and ethanol all reliably reduce the frequency of reticular-elicited hippocampal rhythmical slow activity. In the present experiments we tested a number of drugs which are not usually used for treating generalized anxiety disorders but which have been reported to have some anxiolytic properties. Clonidine (0.3 mg/kg, i.p.), baclofen (6 mg/kg, i.p.) and 8-hydroxy-di-n-propylamino tetralin (8-OH-DPAT) (2.5 mg/kg, i.p.) all reduced the frequency of rhythmical slow activity. The effect of all three drugs was reduced by the 5-hydroxytryptamine 1a antagonist pindolol (2 mg/kg, i.p.). Pindolol had no effect on the reduction in rhythmical slow activity produced by sodium amylobarbitone, as has been previously reported for the benzodiazepine chlordiazepoxide. Flumazenil (10 mg/kg, i.p.), a benzodiazepine receptor antagonist, reduced the effects of chlordiazepoxide (5 mg/kg, i.p.), but not buspirone (10 mg/kg, i.p.). A combination of the selective beta1 adrenergic receptor antagonist metoprolol (20 mg/kg, i.p.) and the beta2 adrenergic receptor antagonist ICI 118,551 (4 mg/kg, i.p.) did not reduce the effects of either buspirone (10 mg/kg i.p.) or diazepam (1 mg/kg, i.p.). These data show that there are at least two separate routes through which anxiolytic agents reduce the frequency of hippocampal rhythmical slow activity. Buspirone, clonidine, baclofen and 8-OH-DPAT act via a system dependent on 5-hydroxytryptamine 1a receptor activation. Benzodiazepines act via activation of the benzodiazepine receptor and probably share with barbiturates action at the GABA-benzodiazepine-chloride ionophore complex but do not produce their effects, directly or indirectly, by 5-hydroxytryptamine 1a receptor activation.