20 



[III. 1NTKKNA1, WdliK OF THE WIND. 



how the action is mechanically realizable in actual practice, but only that it is pos- 

 sible. It is for this purpose, and to understand more exactly thai it can be effected, 

 not only by the process indicated in the second illustration ( Fig. 2), but by another 

 and probably more usual one (and nature has still others at command), that I have 

 considered another treatment of the same conditions of wind-pulsations always 

 moving in the same horizontal direction, but for brief periods interrupted by equal 

 intervals of calm. In this third illustration (Fig. 3) we suppose the body to use 

 the height gained by each pulsation to enable it to descend after each such pulsa- 

 tion, and advance against the direction of the wind. 



Fig. 3. 



The portion A B of the curve represents the path of the plane surface from a 

 state of resl at A, where it has a small upward inclination toward the wind. If a 

 horizontal wind blow upon it in the direction of the arrow, the first movement of 

 the plane will not be in the direction of the wind, but as is abundantly demon- 

 strated by the writer in Experiments in Aerodynamics, it will rise in nearly ver- 

 tical direction, if the angle be small. The wind, continuing to blow in the same 

 direction, at the end of a certain time, the plane, which has risen (owing to its iner- 

 tia and in spite of its weight) to the successive positions shown, is taking up more 

 and more of the horizontal velocitj of the wind, and consequently opposing less 

 resistance to it, and therefore moving more and more laterally, and rising less and 

 less, at every successive instant. 



If the wind continued indefinitely, the plane would ultimately take up its 

 velocity, and finally, of course, fall, when this inertia ceased to oppose resistance to 

 the wind's advance. I have supposed, however, the wind-pulsation to cease at the 

 end of a certain brief period, and, to fix our ideas, let us suppose this period to be 

 five seconds. At this momenl the period of calm begins, and now let the plane, 

 which is supposed to have reached the point 15, change its inclination about a hori- 

 zontal axis t . > that shown in the diagram, falling at first nearly vertically, with its 

 edge on the line of it- descent so as to acquire speed, and this speed, acquired by 

 constantly changing its angle, glide down the curve B C, so that the plane shall be 



