Objective: NaCl solutions (7.5%) have been reported to be effective for resuscitation in animals and trauma patients, but these solutions are not approved for use in the United States. We hypothesized that infusion of Food and Drug Administration-approved 3% NaCl provides superior cardiovascular and metabolic function while reducing the overall fluid requirement for resuscitation of hemorrhage. Our objective was to compare four groups, hypotensive and normotensive resuscitation of hemorrhage using 3% NaCl (HS) or lactated Ringer’s (LR). Design: Sheep were hemorrhaged in three separate bleeds, 25 mL/kg at T0 mins and 5 mL/kg at both T50 and T70 mins. Setting: University laboratory. Subjects: Instrumented conscious sheep. Interventions: Resuscitation was started at T30 mins and continued until T180. Normotensive and hypotensive resuscitation to mean arterial pressures of 90 mm Hg and 65 mm Hg, respectively, was performed with LR or HS using a closed-loop resuscitation system. Measurements and Main Results: All four groups were successfully resuscitated to near target levels. Two animals in the hypotensive treatment protocols died during the second and third bleeding, one with the LR65 group and one with the HS65 group. Mean infused volumes were 59.9 ± 7.0 and 18.0 ± 5.9 in the LR90 and LR65 groups, respectively, and were 19.6 ± 2.2 mL/kg and 13.3 ± 5.7 mL/kg in the HS90 and HS65 treatments (p < .05; LR90 vs. each of the groups). Cardiac indexes were significantly higher with normotensive vs. hypotensive treatment. However, there was no hemodynamic advantage apparent with HS vs. LR when compared with the normotensive or hypotensive treatments. Some animals had high lactate levels (>10 mmol) with both of the hypotensive treatments and also with the HS90 treatment, while not one of the 11 LR90 treatment animals had lactate levels >8 mmol. Conclusions: Volume sparing was apparent with HS, but no hemodynamic or metabolic advantage was apparent when used for either normotensive or hypotensive resuscitation. Trends toward lower base excess values and higher occurrences of deaths only in the hypotensive treatment protocols suggest that resuscitation to a target mean arterial pressure of 65 mm Hg may be too low.