1 This study investigated tachykinin-evoked vasodilatation in the microvasculature of the hamster cheek pouch in vivo. Arterioles and venules were observed by intravital microscopy with video recording, and vasodilatation and constriction, defined as changes in blood vessel diameter, measured by image analysis. All agents were applied topically by superfusion. None of the agents tested had a significant effect on venule diameter. 2 When arterioles were preconstricted (by ca. 50%) with endothelin-1 present in the superfusing medium, substance P (0.3-30 nM) was a potent vasodilator, being 10 fold more active than both neurokinin A and the NK, receptor-selective agonist, substance P methyl ester. The NK2 receptorselective agonist, [P-Ala8]-NKA(4-10)(0.1-10 M) was active only at high concentrations, and the NK3 receptor-selective agonist senktide (O.1 -10 gM) was virtually inactive (n = 8 hamsters). Dilatation evoked by tachykinins and analogues was rapid in onset (<0.5min) and readily reversible. 3 At low concentrations (1-10nM), the non-peptide tachykinin NK, receptor antagonist SR140333.2]octone, chloride) had no effect on the diameter of preconstricted arterioles per se, but potently inhibited dilator responses to substance P methyl ester (apparent pKB 9.9 ± 0.2; n = 5 hamsters, n = 10 estimates). SR140333 (10 nM) did not inhibit submaximal dilator responses evoked by human alpha calcitonin gene-related peptide (aCGRPh; 1.0 nM; P> 0.05; n = 5). 4 The nitric oxide synthase inhibitor NG-nitro-L-arginine methyl ester (L-NAME; 10 gM) caused a 51.3 ± 5.4% arteriolar constriction. In the presence of L-NAME, submaximal vasodilator responses to substance P (10-I00 nM) and carbachol (0.1-1.0 fiM) were significantly attenuated (n = 5 hamsters; P< 0.05) as compared to responses obtained in preparations that were preconstricted to a similar extent by endothelin-l (48.0 ± 5.6%). L-NAME (10 M) was without effect on submaximal vasodilator responses to aCGRPh (0.1 nM) or sodium nitroprusside (1O nM) (n = 5 hamsters; P> 0.05). 5 We conclude that tachykinin-evoked arteriolar vasodilatation in the hamster cheek pouch is mediated via NK, receptor activation and depends, at least in part, on the release of nitric oxide. The NKI receptors mediating vasodilatation can be blocked by topical application of SR140333; which may therefore be useful in the investigation of the role of NK, receptors in neurogenic inflammation in the microvasculature.