The effects of prostaglandin F2 alpha, prostaglandin E1, prostaglandin E2 and the thromboxane A2 analogue U46619 were determined in ring segments of human hand veins. All prostanoids except prostaglandin E1 elicited contraction. The order of potency was U46619 greater than prostaglandin F2 alpha greater than prostaglandin E2. The thromboxane receptor antagonists BM13,505 and AH23848 both caused a parallel rightward displacement of the concentration-response curve for U46619 without depression of the maximum contraction, suggesting competitive antagonism. Schild plots for both antagonists yielded regression lines with slope indices not significantly different from unity. The pA2 values for BM13,505 and AH23848 were 7.9 and 8.4 respectively. Both antagonists also effectively inhibited prostaglandin F2 alpha-induced contractions. However, AH23848 significantly reduced the maximum response, and the results with BM13,505 gave no clear indication of the type of inhibition. In vein segments submaximally contracted by 5-hydroxytryptamine, prostaglandins E1 and E2 produced a biphasic response with a relaxation at low and a contraction at high concentrations. Prostaglandin F2 alpha and U46619 failed to elicit relaxation under these conditions. However, in the presence of either thromboxane receptor antagonist, prostaglandin F2 alpha but not U46619 produced a relaxation. The results are compatible with the presence of at least two prostanoid receptors in human hand veins, a contraction-mediating thromboxane receptor and an as yet unclassified receptor eliciting relaxation. U46619 was a potent agonist at the thromboxane receptor and prostaglandin E1 and E2 preferentially stimulated the relaxation-mediating receptor, whereas prostaglandin F2 alpha appeared to be active at both receptor sites.