Changes in the details of the machine were many and frequent, and after each change there was a flight or an attempted flight. Sometimes we managed to make quite a flight, and others and more numerous merely short "jumps" that would land the machine in a potato patch or a cornfield, where, in the yielding ground, the wheels would crumple up and let the whole thing down. Up to this time we had always used silk to cover the planes, but this proved very expensive and we decided to try a substitute. An entirely new set of planes were made and the new covering put on them. They looked very pretty and white as we took the rebuilt machine out with every expectation that it would fly. Great was our surprise, however, when it refused absolutely to make even an encouraging jump. For a time we were at a loss to understand it. Then the reason became as plain as day; we had used cotton to cover the planes, and, being porous, it would not furnish the sustaining power in flight. This was quickly remedied by coating the cotton covering with varnish, rendering it impervious to the air. After that it flew all right. I believe this was the first instance of the use of a liquid filler to coat the surface cloth. It is now used widely, both in this country and in Europe. We had a great many minor misfortunes with the "White Wing," but each one taught us a lesson. We gradually learned where the stresses and strains lay, and overcame them. Thus, little by little, the machine was reduced in weight, simplified in detail, and finally took on some semblance to the standard Curtiss aeroplane of today.