Bath application of dibutyryl cyclic AMP, 8-bromo cyclic AMP or forskolin to hippocampal slices caused a short-lasting decrease, followed after washout by a long-lasting increase in the amplitude of population spikes recorded in area CA1 in response to 0.05 Hz stimulation of the Schaffer collateral/commissural pathway. Population spike depression lasted only as long as the cyclic AMP analogues were present in the bath, while the potentiation lasted for at least 2 h after they were washed out. Population excitatory postsynaptic potential amplitude was depressed while dibutyryl cyclic AMP was present in the bath, but after washout it was not significantly different from baseline amplitude. The population spike depression caused by dibutyryl cyclic AMP was prevented by the adenosine antagonist 8-p-sulfophenyltheophylline, but the long-lasting potentiation was not. Thus the transient depression induced by dibutyryl cyclic AMP was probably caused by an action of the drug or its breakdown products as adenosine agonists. The long-lasting potentiation is caused by a different mechanism. Occlusion experiments were performed to examine the relationship of the long-lasting population spike potentiation caused by dibutyryl cyclic AMP to the population spike long-term potentiation caused by high-frequency stimulation. High-frequency stimulation delivered after the population spike potentiation caused by dibutyryl cyclic AMP was established did cause a further potentiation, but only up to the level caused by high-frequency stimulation alone. Dibutyryl cyclic AMP administered after high frequency stimulation-induced long-term potentiation was established had no further potentiating effect.(ABSTRACT TRUNCATED AT 250 WORDS)