We used in vivo microscopy and laser-Doppler velocimetry to examine the effects on the gastric mucosal microcirculation and in gastric mucosal blood flow of agents that induce acute gastric mucosal damage. In vivo microscopic observation of superficial mucosal capillaries revealed vascular stasis within a mean of 54, 81, or 61 s after 100% ethanol, 0.6 N HCl, or 0.2 N NaOH, with the subsequent development of hemorrhagic mucosal lesions. Mucosal blood flow estimated by laser-Doppler velocimetry decreased by 30% at 5 min after luminal application of 100% ethanol, and decreased further to about 40% of basal levels by 15 min. The decreased mucosal blood flow 15 min after application of 50% ethanol correlated with the extent of hemorrhagic mucosal lesions. Examination of the submucosal vessels that supply and drain the mucosa showed moderate dilation of small arterioles 1, 3, and 6 min after exposure to 100% ethanol but there were no consistent changes in venules. Mild vasoconstriction of small- and medium-sized venules could be detected 6, 10, and 15 min after NaOH but not after exposure to HCl. Pretreatment with 16,16-dimethyl prostaglandin E2 or sodium thiosulfate before exposure of the mucosa to ethanol prevented capillary stasis, maintained mucosal blood flow, and prevented the development of hemorrhagic gastric mucosal lesions. Topical mucosal application of 16,16-dimethyl prostaglandin E2 decreased, whereas topical exposure to sodium thiosulfate increased gastric mucosal blood flow, indicating that change in blood flow per se is an unlikely mediator of protection.