5-Hydroxytryptamine (5-HT) receptor-mediated contraction of endothelium denuded rabbit middle (MCA) and posterior (PCA) cerebral arteries was characterized by use of selective agonists and antagonists for different 5-HT receptor subtypes. 2. 5-HT and various 5-HT receptor agonists contracted the arteries with the following rank order of potency in MCA: 5-carboxamidotryptamine (5-CT) > 5-HT > 5-methoxytryptamine (5-MeOT) > sumatriptan > alpha-methyl-5-HT (alpha-Me-5-HT) >> 8-hydroxy-2-(di-n-propylamino) tetralin (8-OH-DPAT) and in PCA: 5-CT > 5-HT > sumatriptan > 5-MeOT > alpha-Me-5-HT >> 8-OH-DPAT. With few exceptions, the maximal contractile responses of these agonists were similar to that induced by 5-HT. 3. The selective antagonists of 5-HT2A/2C (ketanserin), 5-HT4 (SDZ 205-557) and 5-HT1A/1B (S-(-)-propranolol) sites were devoid of inhibitory effect on 5-HT-mediated contraction in both MCA and PCA, thus excluding activation of the corresponding receptors. 4. In both arteries, the contraction-response curve to 5-HT was unaffected by the 5-HT3 receptor antagonist, ICS 205-930 (0.01 and 0.1 microM) whilst a small (3 and 6 fold displacement) was seen with MDL 72222 (0.1 and 1 microM). 5. The mixed 5-HT1-like/5-HT2A receptor antagonist, methiothepin (0.001-0.1 microM), was a potent antagonist of 5-HT-induced contractions in both arteries, giving pA2 values of 9.4 +/- 0.7 and 9.6 +/- 0.8 in MCA and PCA, respectively. 6. Rauwolscine (O.1-10 MicroM) and yohimbine (0.3, 3 MicroM) inhibited contractions to 5-HT in a competitive manner, pA2 values of 7.1 +/- 0.6 and 6.7 +/-0.6 were determined for rauwolscine in MCA and PCA,respectively. An apparent pA2 value of 6.9 +/-0.2 was calculated for yohimbine (3 MicroM) in both MCA and PCA.7. In conclusion, these results suggest that the contractile response to 5-HT in rabbit isolated MCA and PCA is predominantly mediated by the 5-HTID receptor subtype, although a small contribution by 5-HT3 receptors cannot be excluded.