Since the road-service tests of the four special fuels supplied by the Research Department, made under 1922 summer-weather conditions, gave results that were deemed inconclusive, arrangements were made for a repetition of both series of tests under the winter-weather conditions of 1923 to determine whether the relative fuel mileages for different fuels are dependent on the temperature at which car operation is conducted. The paper is a report upon the results obtained.
Four fuels that bore a relation to those used in the 1922 summer tests were specified and means adopted whereby knowledge of their quality was concealed from the drivers, special emphasis being placed on crankcase-oil dilution and on the performance as reflected by the drivers' comments. The method of the test was, for the most part, that of the summer of 1922, with cars operated as usual by their regular drivers, the filling with fuel and oil and the keeping of records being wholly in charge of the technical staff of the company concerned.
The details of procedure are presented, together with voluminous tabular data, and comment is made. A consistent and important decrease in mileage for the winter tests is apparent. The conclusions reached are stated in detail, as a summary, under six divisions.