The heaviest flying birds known scarcely reach one quarter of the weight of an average man, and it can be said that one of the most probable causes of the absence in Nature of heavier varieties is the fact that although such birds could fly easily enough, once on tlie wing, they would be unable to leave the ground, owing to’ the enormity of the power the first lift Avould require. The larger and heavier the bird, the less, in relation to its weight, is the power developed in flight, but, on the other hand, the greater is the eft'ort. needed for a. lift from the ground. I do not wish to enlarge upon this question, which has been treated by various authors, and only refer to it because it embodies a principle which we must bear in mind when considering the means of starting flight, namely, that the difference between the power required respectively for the flight and for the start becomes greater as the apparatus increases in weight.

The fact that a considerable amount of knowledge can lie gained by experiments with weighted model gliders does not at present appear to be sufficiently appreciated ; only with model gliders, for convenience sake, a less weight is used per square foot of carrying aëroplane surface. A ¼ lb. per square foot area is a convenient demonstrating weight, giving a Speed of about 8 m.p.h., with a drop of about one in six ; with man-carrying machines the usual weight varies between ½ and 1½ lbs.

If a single aëroplane be released in the air it will simply fall anyhow to the ground, but on fixing a steering plane fore or aft, let us say in front as in Fig. 1, there will be a tendency for it to glide, but sooner or later one side will dip downwards, owing to the two planes not being in the same straight line or being improperly balanced.

This paper is intended in the first place to ascertain some definite and tangible ideas as to dimensions of the vital parts of machines “ heavier than air” designed for aërial navigation. The vital parts are, of course, the engine and propeller.

In the second place it is proposed to discuss the aëroplane and its functions in an airship.

In all engineering problems the first requisite towards a solution is to reduce things to numerical values. An inventor may imagine a design for a machine, but he cannot put it into practice without calculating out the necessary dimensions, and such, calculations, when undertaken, often either confirm or condemn the design.

Much of the experimenting at present going on is evidently entered upon, and carried on, without any calculations at all, the only results being a waste of time and money.