The quantitative [14C]2-deoxyglucose autoradiographic technique was applied to the measurement of the cerebral metabolic effects of adenosine A1 and A2 receptor agonists and antagonists in adult rats. The adenosine A1 receptor agonist and antagonist, 2-chloro-N6-cyclopentyladenosine (CCPA) and 8-cyclopentyl-1,3-dipropylxanthine (DPCPX) as well as the adenosine A2 receptor agonist, 2-[p-(2-carboxyethyl)phenylethylamino]-5'-ethylcarboxamidoadenosin e (CGS 21680), were injected at the dose of 0.01 mg/kg. The adenosine A2 receptor antagonist, 3,7-dimethyl-1-proparglyxanthine (DMPX) was injected at the dose of 0.3 mg/kg. These doses were chosen in accordance with the known affinity of the drugs for their respective receptor and to avoid peripheral effects. The adenosine A1 receptor agonist, CCPA, induced decreases in glucose utilization in three brain areas, the globus pallidus and two hypothalamic nuclei. The adenosine A2 receptor agonist, CGS 21680, induced more general depressant effects on energy metabolism which were significant in 17 brain areas, such as cerebral cortex, hippocampal and white matter regions plus motor and limbic structures. The adenosine A2 receptor antagonist, DMPX, decreased glucose utilization in the globus pallidus while increasing energy metabolism in the cochlear nucleus. The adenosine A1 receptor antagonist, DPCPX, depressed glucose utilization in the globus pallidus and dentate gyrus, and increased rates of energy metabolism in six regions, mainly hypothalamic, thalamic areas and in the cochlear nucleus. There was a mismatch between cerebral metabolic consequences of adenosine A1 and A2 receptor agonists and the localization of corresponding adenosine receptors. The metabolic effects of the adenosine A2 receptor agonist and antagonist were consistent with the known involvement of that type of receptor in the control of locomotion and its effects on neuronal firing in the hippocampus and cerebral cortex. The effects of the adenosine A1 receptor agonist were very discrete and mostly related to the transient decrease in blood pressure induced by the drug. The increases in glucose utilization induced in limbic regions by the adenosine A1 receptor antagonist are probably linked to the regulation by adenosine of arousal and cardiorespiratory function. These results are in good agreement with the neuroregulatory function of the adenosine system as previously shown by other methods.