We placed our first orders for cotton airplane fabric in September, 1917—orders for 20,000 yards—and from that time on the use of linen decreased. By March of 1918 the production of cotton airplane cloth had reached 400,000 yards per month. In May the production was about 900,000 yards; and when the war ended this material was being turned out at the rate of 1,200,000 yards per month. Starting with a few machines, our cotton mills had gradually brought 2,600 looms into the enterprise, each loom turning out about 120 yards of cloth in a week. A total of 10,248,355 yards of cotton fabric was woven and delivered to the Government—over 5,800 miles of it, nearly enough to reach from California to France. The use of cotton fabric so expanded that in August, 1918, we discontinued the importations of linen altogether.
There was, however, danger that we would be limited in our output of cotton fabric if there were any curtailment in the supply of the long-staple sea-island and Egyptian cotton of which this cloth is made. To make sure that there would be no shortage of this material the Signal Corps in November, 1917, went into the market and purchased 15,000 bales of sea-island cotton. This at all times gave us an adequate reserve of raw material for the new fabric.
Cotton proved to be not only an admirable substitute for linen but even a better fabric than the original cloth which had been used. No matter how abundant the supply of flax may be, it is unlikely that linen will ever again be used in large quantities for the manufacture of airplane wings.
Thus, as the airplane situation was saved by the prompt action of the Signal Corps in organizing and training the spruce industry, so again its decision to produce cotton fabric and its prompt action in cornering the necessary cotton supply made possible the uninterrupted expansion of the allied aviation program.
The wings of an airplane must not only be covered with fabric, but the fabric must be filled with dope, which is a sort of varnish. The function of the dope is to stretch the cloth tight and to create on it a smooth surface. After the dope is on the fabric the surface is protected further by a coat of ordinary spar varnish.
We found in the market two sorts of dope which were being furnished to airplane builders of all countries by various chemical and varnish manufacturers. One of these dopes was nitrate in character and was made from nitrocellulose and certain wood-chemical solvents including alcohol. This produced a surface similar to that of a photographic film. The other kind of dope had an acetate base and was made from cellulose-acetate and such wood chemical solvents as acetone.
The nitrate dope burned rapidly when ignited but the acetate type was slow burning. Thus in training planes not subject to attack by enemy incendiary bullets the nitrate dope would be fairly satisfactory, but in the fighting planes the slow-burning acetate dope was a vital necessity. Up to our participation in the war the dopes produced in the United States were principally nitrate in character.
It was evident that we should make our new dope acetate in character to avoid the danger of fire. But for this we would require great quantities of acetone and acetate chemicals, and a careful canvass of the supply of such ingredients showed that it would be impossible for us to obtain these in anything like the quantities we should require without developing absolutely new sources of production.
Already acetone and its kindred products were being absorbed in large quantities by the war production of the allies. The British Army was absolutely dependent upon cordite as a high explosive. Acetone is the chemical basis of cordite; and therefore the British Army looked with great concern upon the added demand which the American aviation program proposed to put upon the acetone supply.