GLIDING 17 
Very fast flyers, such as the Swift, have very little 
concavity. There is another fact, too, that points 
to the same conclusion: in a bird’s wing the part 
that is near the body is the most concave, while 
towards the extremity it becomes nearly flat, in 
some cases quite flat. Now the extremity of the 
wing, when a stroke is given, moves with great 
velocity forward as well as downward, the near 
part comparatively slowly. Thus birds seem to be 
exponents of the principle that the curve should 
vary inversely as the speed. And here I must 
point out a great advantage the bird has over his 
upstart rival, the aeroplane. The aeroplane is 
rigid, with at most the rear edge only of its plane 
flexible. If the principle to which I have called 
attention is sound, it must be built either with a 
curve appropriate to great speed or with one which 
Suits comparatively slow travelling. It cannot 
be varied during flight according as the pace is 
varied. But when the bird takes a very hard 
stroke, the front-to-back curve of his wings, which 
is mainly the curve of the feathers, is much reduced, 
so that by plying his wings with great vigour he to 
some extent modifies his configuration and adapts 
it to his high ey (For birds in rapid flight, 
see Pls. Iv-vII.) 
It is always well to ask “‘ Why is it so ?” when- 
ever experiment discovers a fact. Why, then, 
with a curved surface is the lift more and the drift 
less? It is very difficult to see how the air does 
its work, and theory on such a subject may prove 
to be only random guesswork. But the lift would 
seem to be greater from the simple fact that the 
Cc 
