Blends of up to 20% methanol in gasoline were evaluated in both engine dynamometer and controlled vehicle tests, and in a 50,000 mile road test. Performance comparisons between methanol blends and base gasolines were made in vehicle driveability and vapor lock tendency, engine deposits and wear, fuel economy, exhaust emissions, compatibility with fuel system materials, and phase stability of the blends. Vapor lock tests in six 1974 cars strongly suggested that the vapor lock tendency of methanol blends is greater than would be predicted for gasolines having the same volatility characteristics. Cold start and warm-up driveability of two 1974 cars at 70°F depreciated as methanol concentration increased in base fuels of three volatility levels. These driveability data were found to correlate well, at a given methanol concentration, with fuel volatility characteristics described by means of a new fuel vaporization pressure technique. Engine deposits, rusting, wear and crankcase oil performance were not significantly affected by methanol in gasoline when evaluated in four laboratory engine tests and in the four-car, 50,000 mile road test Fuel consumption was found generally to increase with increasing methanol concentration, while exhaust emissions from both vehicles and stationary engines varied as expected from the stoichiometric air/fuel ratio of the blends. With the possible exception of copper and brass, screening tests with methanol blends indicated no serious potential compatibility problems with the fuel system metals tested or with Buna N or Neoprene rubbers. Methanol solubility in gasoline was shown to be increased by co-solvent alcohols. These also increased the stability of blends to phase separation by added water, but still phase separation occurred with as little as 0.1 or 0.2% of water.