The effects of 2-chloro-N(6)-cyclopentyladenosine (CCPA) perfused intrahippocampally (1 microM) and injected intraperitoneally (0.5 mg/kg) were investigated in focally-evoked pilocarpine-induced (10 mM) seizures in freely moving rats. While the intrahippocampal perfusion of this highly selective adenosine A(1) receptor agonist gave complete protection against pilocarpine-induced seizures, systemic administration only partially protected the animals, as evaluated by concomitant behavioural and electrocorticographical (ECoG) observations and monitoring of the neurotransmitter alterations. However, pilocarpine-evoked elevation of hippocampal glutamate overflow was significantly attenuated by CCPA irrespective of the mode of administration. Acute pretreatment with systemic 8-cyclopentyl-1,3-dipropylxanthine, a selective A(1) antagonist, reversed both the partial protective effect and the attenuating effect on the extracellular glutamate elicited by systemic CCPA administration. Intrahippocampal CCPA markedly reduced basal hippocampal dopamine efflux but not GABA or glutamate and considerably attenuated the pilocarpine-evoked elevation in dopamine levels. Systemic CCPA appeared to have little influence on the overall pattern of dopamine elevation. The findings give evidence that CCPA-elicited abatement of the evoked glutamate release alone, cannot fully account for its anticonvulsant effect and may suggest that the effects mediated by adenosine on postsynaptic adenosine receptors could be more crucial for its anticonvulsant effect.