How he overcame the various difficulties that faced him and constructed a steam-engine capable of the task allotted to it forms a story in itself, too long for recital here. His first power-driven aerodrome of model size was begun in November of 1891, the scale of construction being decided with the idea that it should be large enough to carry an automatic steering apparatus which would render the machine capable of maintaining a long and steady flight. The actual weight of the first model far exceeded the theoretical estimate, and Langley found that a constant increase of weight under the exigencies of construction was a feature which could never be altogether eliminated. The machine was made principally of steel, the sustaining surfaces being composed of silk stretched from a steel tube with wooden attachments. The first engines were the oscillating type, but were found deficient in power. This led to the construction of single-acting inverted oscillating engines with high and low pressure cylinders, and with admission and exhaust ports to avoid the complication and weight of eccentric and valves. Boiler and furnace had to be specially designed; an analysis of sustaining surfaces and the settlement of equilibrium while in flight had to be overcome, and then it was possible to set about the construction of the series of model aerodromes and make test of their 'lift.' By the time Langley had advanced sufficiently far to consider it possible to conduct experiments in the open air, even with these models, he had got to his fifth aerodrome, and to the year 1894. Certain tests resulted in failure, which in turn resulted in further modifications of design, mainly of the engines. By February of 1895 Langley reported that under favourable conditions a lift of nearly sixty per cent of the flying weight was secured, but although this was much more than was required for flight, it was decided to postpone trials until two machines were ready for the test. May, 1896, came before actual trials were made, when one machine proved successful and another, a later design, failed. The difficulty with these models was that of securing a correct angle for launching; Langley records how, on launching one machine, it rose so rapidly that it attained an angle of sixty degrees and then did a tail slide into the water with its engines working at full speed, after advancing nearly forty feet and remaining in the air for about three seconds. Here, Langley found that he had to obtain greater rigidity in his wings, owing to the distortion of the form of wing under pressure, and how he overcame this difficulty constitutes yet another story too long for the telling here.