Our study examined the efficacy of four treatment modalities in controlling hemorrhage and achieving hemodynamic stabilization in hemorrhagic shock: intravenous fluid replacement (IV); military antishock trousers used concomitantly with fluids (MAST); balloon occlusion at the level of the diaphragm with concomitant fluid replacement (balloon); and a combination of MAST inflation, balloon occlusion, and fluid resuscitation (MAST and balloon). Twenty-eight mongrel dogs were anesthetized, and the spleen was exposed and completely crushed. The abdomen was closed, and treatment was initiated and continued for four hours or until the dog died. For all conditions the hematocrit dropped during the course of the experiment; balloon occlusion was effective at slowing this drop (P less than .0001), but MAST had no statistically significant effect. Animals with balloons bled more slowly into the abdominal cavity than did animals in the other two groups (P less than .0001). MAST also were effective at slowing the bleeding (P less than .05). Of the balloon and the MAST and balloon dogs, all except one survived the entire four hours; this difference between balloon and nonballoon dogs is significant (P = .002). MAST did not have a statistically significant effect on survival. Perfusion pressure (PP) declined during the course of the experiment, and the balloon was effective at slowing this decline (P less than .0001); none of the other comparisons was statistically significant.