1. The relative contribution of ETA and ETB receptors in the response of rat skin to endothelins was investigated by use of the selective ETB agonist IRL-1620 and the selective ETA antagonist BQ-123. 2. Binding data suggest the presence of ETA and ETB receptors as preincubation with [Ala3,11,18Nle7]-endothelin-1 reduced ET-1 binding by approximately 40%. 3. Intradermal injection of endothelin-1 (ET-1, 1-10 pmol/site) and ET-3 (3-100 pmol/site) induced a dose-dependent decrease in local blood flow assessed by 133Xe clearance at test sites in rat skin. 4. The endothelin analogue [Ala3,11,18Nle7]-ET-1 (30-1000 pmol/site) induced significant vasoconstriction (P < 0.05) at the highest doses used and the selective ETB receptor agonist, IRL-1620 [Suc[Glu9,Ala11,15] endothelin (8-21)], (0.01-100 pmol/site) acted in a potent manner to induce a significant (P < 0.01) dose-dependent decrease in 133Xe clearance. 5. Co-injection with the selective ETA receptor antagonist, BQ-123 (1 nmol/site), completely abolished the vasoconstriction to ET-1 and partially to ET-3, but had no effect on IRL-1620-induced vasoconstriction. In addition, IRL-1620 responses were not altered at sites treated with submaximal doses of a nitric oxide synthase inhibitor or a prostaglandin synthase inhibitor. 6. ET-1 and IRL-1620 (100 fmol-1 pmol/site) did not induce oedema formation as measured by [125I]-albumin accumulation in the presence or absence of the vasodilator, calcitonin gene-related peptide (CGRP). ET-1 (1-3 pmol/site) inhibited substance P-induced oedema formation and this effect,suggested to be secondary to a vasoconstrictor effect, was significantly reversed by BQ-123 (1 nmol/site).7. The findings in this study indicate that there are ETA and ETB receptors in rat skin and agents which activate either receptor act to mediate a decrease in cutaneous blood flow, but have no effect on increased microvascular permeability.