The paper is a report covering an investigation made by the Bureau of Standards to secure data that can be used as a basis for estimating the effect of a change in gasoline volatility on the fuel consumption of cars now in service throughout the United States. Actual tests began in August, 1922, to determine the effect of four fuels of different characteristics on the number of car-miles obtainable per gallon of fuel, as well as on crankcase-oil dilution.
Descriptions are given of the fuels used, the test-cars, the apparatus, the tests and the test methods, inclusive of the crankcase-oil dilution investigation, voluminous tabular data accompanying the text. Table 12 gives a summary of the results and these are discussed briefly. The program for a proposed similar investigation under low-temperature conditions is outlined.