1. A range of naturally-occurring prostaglandins sulprostone, 16,16-dimethyl prostaglandin E2 (DME2) and the thromboxane A2 (TXA2)-mimetic, 11 alpha,9 alpha-epoxymethano prostaglandin H2 (U-46619) have been tested for contractile agonist activity on human isolated bronchial smooth muscle. 2. Prostaglandin D2 (PGD2), PGF2 alpha, 9 alpha,11 beta-PGF2 (11 beta-PGF2) and U-46619 all caused concentration-related contractions. U46619 was at least 300 fold more potent than the other prostanoids with a mean EC50 of 12 nM. Sulprostone caused contraction only at the highest concentration tested (30 microM). PGE2 and PGI2 caused relaxations at low concentrations, and only caused contractile responses at high concentrations (greater than or equal to 10 microM). In contrast, DME2 caused small contractions at low concentrations but relaxation at the highest concentration tested (30 microM). 3. The rank order of contractile agonist potency was: U-46619 much greater than 11 beta-PGF2 congruent to PGF2 alpha greater than PGD2 greater than PGE2 greater than PGI2 congruent to sulprostone congruent to DME2. 4. The TP-receptor blocking drug, AH23848 (1 microM) antagonized the contractile effects of U-46619, PGD2, PGF2 alpha and 11 beta-PGF2, but had no effect against contractions to carbachol. In a single experiment, a pA2 of 8.3 (slope = 1.2) was obtained for AH23848 against U-46619. 5. In most preparations, administration of AH23848 (1 microM) to human bronchus resulted in small, transient contractile responses. 6. The results obtained with both the agonists and the antagonist, AH23848 are therefore consistent with prostanoid-induced contractions of human bronchial smooth muscle being mediated by TP-receptors.