At this point in the development we were not aware of the great production of spare parts that would be necessary. Yet we did understand that there must be a considerable production of spares; and in order to take the burden of this manufacture from the regular airplane plants, and also to educate other factories up to the point where they might undertake the construction of complete airplanes, we placed many contracts for spare parts with widely scattered concerns. Among the principal producers of spares were the following: The Metz Co., Waltham, Mass. Sturtevant Aeroplane Co., Jamaica Plains, Mass. Wilson Body Co., Bay City, Mich. West Virginia Aircraft Corporation, Wheeling, W. Va. The Rubay Co., Cleveland, Ohio. Engel Aircraft Co., Niles, Ohio. Hayes-Ionia Co., Grand Rapids, Mich. For a long time the supply of spare parts was insufficient for the needs of the training fields. This was only partly due to the early lack of a proper realization of the quantity of spares that would be required. The production of spares on an adequate scale was hampered by numerous manufacturing difficulties incident to new industry of any sort in shops unacquainted with the work, and by a lack of proper drawings for the parts. As to the training planes themselves, with all factories in the country devoting themselves to this type exclusively at the start, the production soon attained great momentum. The Curtiss Co. particularly produced training planes at a pace far beyond anything previously obtained. The maximum production of JN-4 machines was reached in March, 1918, when 756 were turned out. Advanced training machines are faster, traveling at about 105 miles per hour; and they carry various types of equipment to train observers, gunners, photographers, and radio men. For this machine we adopted the Curtiss JN-4H, which was substantially the same as the primary training plane, except that it carried a 150-horsepower Hispano-Suiza engine. We also built a few "penguins," a kind of half airplane that never leaves the ground; but this French method of training with penguins we never really adopted. The finishing school for our aviators was in France, where the training was conducted in Nieuports and other fighting machines. In July, 1918, we reached the maximum production of the advanced training machines, the output being 427. As the supply of primary training planes met the demands of the fields, the production was reduced, since the original equipment, kept up by only enough manufacture to produce spares and replacement machines, would suffice to train all of the aviators that we would need. The actual production of training planes by months was as follows: