A prospective study was undertaken to evaluate the change in hematocrit produced by infusion of crystalloid solution. Twenty healthy, nonbleeding subjects were randomized into two groups. Each group had an initial, baseline hematocrit. Group A then received an infusion of 20 mL/kg normal saline over 45 minutes and had a second hematocrit drawn, after which an additional 15 mL/kg was infused over one hour and a third hematocrit was drawn. Group B also had an initial 20 mL/kg infusion over 45 minutes followed by a second hematocrit determination, but this was followed by an additional 15 mL/kg over three hours, again followed by a final hematocrit. Multiple methodologic factors were included to stimulate, as closely as possible, an actual trauma treatment protocol. In each group, a significant decrease in hematocrit was produced by the first infusion, specifically, 4.8 points in group A and 4.1 points in group B, with more than 50% of the subjects demonstrating a drop of 5 points or more. The second infusion produced no further significant change in either group. There was no difference in the magnitude of the change between the two groups. The study demonstrates that a significant drop in hematocrit may be expected in nonbleeding patients who receive crystalloid infusions during trauma evaluation protocols.