&NA;Small mesenteric arteries supplying partially isolated jejunal segments were totally occluded for 5 minutes and then released. With video microscopy, blood flow was calculated from measurements of submucosal arteriolar diameter and red blood cell velocity. For the first 30 minutes of reperfusion, the serosa was superfused with a Ringer's vehicle containing either adenosine (ADO; 10‐4M), acetylcholine (ACh; 10‐5M), or prostacyclin (PGI2; 3×10‐7M). Thereafter, the substances were removed from the suffusate, and superfusion continued with vehicle alone for an additional 10‐30 minutes. These concentrations were equieffective for causing vasodilation. During the first minute of reperfusion, blood flow increased more than 300% of baseline in all groups. Within the subsequent 30 minutes, blood flow fell to 45±3% of baseline with vehicie alone, which demonstrates the no‐reflow phenomenon. While either ADO, ACh, or PGI2was in the suffusate, vasodilation was persistent. After washout of these substances, the postocclusion blood flows were significantly higher with each treatment than with vehicle alone, which shows that each substance had a positive action. However, with ADO, blood flow was 121±7% of baseline after washout, whereas with ACh or PGI2, it was 64±10% or 69±5% of baseline after washout. This property of ADO was observed if the mucosa was superfused with a Ringer's solution or with a bile salt solution, which suggests that ADO might have similar properties in situ. After 60 minutes of reperfusion, the intestinal villi were short, thick, and edematous with epithelial necrosis and crypt degeneration. ADO attenuated most of these histological changes to a greater extent than either PGI2or ACh. Furthermore, ADO reduced a biochemical index of neutrophil infiltration; tissue myeloperoxidase concentration was increased to 169±14% of baseline with vehicle but was increased to 120±8% with ADO. Overall, these observations suggest that ADO protects the intestine from ischemia‐reperfusion injury by causing vasodilation and by inhibiting neutrophil function. The vasodilatory effect probably is a minor component because other vasodilators (ACh and PGI2) had minimal protective effects in these conditions. (Circulation Research1989;65:426‐435)