This study compares the exhaust emission effects of splash blending MTBE with the effects of splash blending an equal volume of a “hydrocarbon surrogate” of similar volatility. A commercially available mixture of hexane isomers (C6-isomerate) served as the “hydrocarbon surrogate.” Three fuels were examined: an industry-average base gasoline, the base gasoline splash blended with 15 vol % MTBE, and the base gasoline splash blended with 15 vol % of C6-isomerate. Both criteria and speciated exhaust emissions were measured using the Federal Test Procedure in a fleet comprised of ten 1989-1993 model year (MY) vehicles.The only difference observed in the criteria and toxic emissions for the two blending agents was the level of CO emissions. The addition of MTBE decreased CO by 13% relative to the base fuel while the hydrocarbon surrogate resulted in a 5% decrease. Although the effect of MTBE was statistically significant, the effect of the surrogate was not. For emissions of total hydrocarbon, NOx, and toxics, the effects observed for either of the blending agents were equivalent. In the case of total hydrocarbon, both blending agents yielded a statistically significant reduction of about 10% relative to base. Small decreases in NOx emissions were observed for both blending agents, but they were not statistically significant. These results imply that, for this vehicle fleet, the only beneficial change that can be attributed specifically to the presence of oxygen is a reduction in CO emissions.