Radioactive microspheres were used to measure blood flow in the cat stomach during exposure to saline, 0.075 M HCl, and then 15 and 30 min after exposure to 20 or 40 mM aspirin in HCl. At the end of the experiment, the stomach wall was divided into ulcerated regions and adjacent nonulcerated areas. When exposed to saline, both regions had similar blood flow: 27 +/- 5 and 25 +/- 5 ml.min-1.100 g-1 (means +/- SE). Addition of acid caused a significant increase in blood flow to 41 +/- 7 ml.min-1.100 g-1 only at those sites that eventually ulcerated in the presence of aspirin. In the adjacent nonulcerated regions, blood flow was 31 +/- 5 ml.min-1.100 g-1 and was not significantly greater than the flow recorded during saline exposure. Aspirin caused ulcer site blood flow to increase dramatically to 89 +/- 12 and 122 +/- 18 ml.min-1.100 g-1 after 15 and 30 min, whereas the adjacent nonulcerated tissue rose to 40 +/- 6 and 44 +/- 5 ml.min-1.100 g-1, respectively. The ulcer site hyperemia with acid alone suggests higher mucosal permeability in these regions allowing back-diffusion of acid and injurious agents. The present data obtained in the cat do not support the notion that ischemia plays a role in initiating nonsteroidal anti-inflammatory drug (NSAID)-induced ulcers, but rather that acute NSAID ulcers are associated initially with a hyperemia.