1 The effects of several K+ channel blockers (sulphonylureas, 4-aminopyridine and tetraethylammonium) on the antinociception induced by clonidine, baclofen and U50,488H were evaluated by use of a tail flick test in mice. 2 Clonidine (0.125-2 mg kg-1, s.c.) induced a dose-dependent antinociceptive effect. The ATP-dependent K+ (K(ATP)) channel blocker gliquidone (4-8 mug/mouse, i.c.v.) produced a dose-dependent displacement to the right of the clonidine dose-response line, but neither 4-aminopyridine (4-AP) (25-250 ng/mouse, i.c.v.) nor tetraethylammonium (TEA) (10-20 mug/mouse, i.c.v.) significantly modified clonidine-induced antinociception. 3 The order of potency of sulphonylureas in antagonizing clonidine-induced antinociception was gliquidone>glipizide>glibenclamide>tolbutamide, which is the same order of potency as these drugs block K(ATP) channels in neurones of the CNS. 4 Baclofen (2-16 mg kg-1, s.c.) also induced a dose-dependent antinociceptive effect. Both 4-AP (2.5-25 ng/mouse, i.c.v.) and TEA (10-20 mug/mouse, i.c.v.) dose-dependently antagonized baclofen antinociception, producing a displacement to the right of the baclofen dose-response line. However, gliquidone (8-16 mug/mouse, i.c.v.) did not significantly modify the baclofen effect. 5 None of the K+ channel blockers tested (gliquidone, 8-16 mug/mouse; 4-AP, 25-250 ng/mouse and TEA, 10-20 mug/mouse, i.c.v.), significantly modified the antinociception induced by U50,488H (8 mg kg-1, s.c.). 6 These results suggest that the opening of K+ channels is involved in the antinoceptive effect of alpha2 and GABA(B), but not kappa-opioid, receptor agonists. The K+ channels opened by alpha2-adrenoceptor agonists seem to be ATP-dependent channels, whereas those opened by GABA(B) receptor agonists are not.