In regional lung injury, pulmonary blood flow decreases to the injured regions, and anastomotic bronchial blood flow and total bronchial blood flow increase. However, the pattern of redistribution of the two blood flows to the injured and noninjured areas is not known. In six anesthetized sheep, pulmonary and bronchial blood flows were measured with 15-mu m fluorescent microspheres by using the reference flow method. Blood flows were measured in the control state and 1 h after instilling 1 ml/kg of 0.1 N hydrochloric acid into a dependent segment of the left lung. The lungs were then removed, dried, and cubed into similar to 2-cm cubes while spatial coordinates were noted. Blood flow to each piece was calculated. Mean pulmonary blood flow to the noninjured pieces went from 730 +/- 246 to 574 +/- 347 ml/min (P = 0.22), whereas in the injured pieces the pulmonary blood flow decreased from 246 +/- 143 to 56 +/- 46 ml/min (P < 0.01). In contrast, bronchial blood flow to the injured pieces increased from 0.51 +/- 0.1 to 1.43 +/- 0.85 ml/min (P = 0.005). We measured the change in flow as it related to the distance from the center of the injured area. Pulmonary blood flow decreased most at the center of the injury, whereas bronchial blood flow doubled at the center of injury and decreased with the distance away from the injury. The absolute increase in bronchial blood flow was substantially less than the decrease in pulmonary blood flow in the injured pieces. We also partitioned the observed variation in pulmonary and bronchial blood flow into that attributable to structure and that due to lung injury and found that 48% of the variation in pulmonary blood flow could be attributed to structure, whereas in the bronchial circulation 70% was attributable to structure. The reasons for these differences are not known and may reflect the intrinsic properties of the systemic and pulmonary circulations.