1. The purpose of the present study was to compare the effects of several alpha-adrenoceptor agonists and antagonists on cerebral cortical overflow of endogenous noradrenaline (NA) in freely moving rats. One or two days after the implantation of transcerebral dialysis tubes in the frontoparietal cortex, extracellular NA levels were monitored on-line with high performance liquid chromatography and electrochemical detection. The drugs were applied locally via the dialysis membrane, and effects on NA overflow were determined in cumulative concentration-response curves. 2. The average basal cortical NA overflow of all experiments was 0.25 pg min-1. The alpha 2-adrenoceptor agonists caused a concentration-dependent decrease in NA levels. UK-14,304 was the most potent and B-HT 933 the least potent agonist. The maximal decrease in NA overflow was to 10-15% of control levels after UK-14,304 or moxonidine, to 30% after clonidine and to 50% after B-HT 933 administration. Continuous activation of the presynaptic alpha 2-adrenoceptor with 10(-6) M UK-14,304 caused a decrease in NA levels to 40-50% of basal levels. This decrease was reached within 1 h and remained stable for the entire 3 h measurement period. The alpha 1-adrenoceptor agonists, phenylephrine and methoxamine, induced an increase in NA levels to 225% and 300%, respectively, at a concentration of 10(-3) M. 3. Local application of alpha 2-adrenoceptor antagonists caused an increase in NA levels, with idazoxan being more potent than piperoxan. Yohimbine did not cause any significant change. 4. All drugs used in these in vivo experiments had in vitro recoveries across the dialysis membrane between 10 and 20%. 5. We conclude that microdialysis with local drug application is suitable for the comparison of the pharmacological effects of drugs with affinity for alpha-adrenoceptors on cortical NA overflow in vivo,provided that the passage across the membrane is equal for the different drugs.