In the stability investigations which the late Captain Ferber published in the Revue d'Artillerie, the sustaining and other surfaces of an aeroplane were in certain cases taken to be represented, for dynamical purposes, by a system of three plane resisting laminæ fixed mutually at right angles. Unfortunately, however, such a system cannot in general be made equivalent to a collection of surfaces, such as those of an aeroplane, with the result that Captain Ferber's investigation failed to give the correct conditions of lateral stability. At the same time, Ferber's system of three orthogonal planes is so convenient, especially for forming a general idea of the effects of wind gusts on an aeroplane, that it is desirable to investigate conditions and limitations under which such a representation is valid. The desirability of a further investigation of the forces and couples acting on a system of resisting surfaces of a general character was foreshadowed in “Stability in Aviation,” and a more detailed discussion of the problem has now become necessary in order to prepare the way for further studies in the rigid dynamics of the motions of aeroplanes or of systems resembling them.

The first direct statement of the problem of controlling an aeroplane, without the permanent exercise of physical effort by the pilot, appears in this paper. During the state of transition from straight flying to flying in a circle, a pilot exerts a force to pull over the rudder and in the usual case it is necessary to maintain some force in order to make a continuous turn. Professor Bryan proposes the design of an aeroplane such that turning having been started, maintenance of turning is obtained without any effort being necessary to keep the rudder in position. So far as the pilot is concerned the same result would be obtained by balancing the control surfaces, but the effect on the aeroplane is quite different. In the case of balanced controls a lateral force is still exerted by the rudder and so stresses are introduced in the tail structure of the aeroplane ; on the other hand, in an inherently controllable aeroplane, there is no lateral force on the rudder when turning and, therefore, no stresses in the tail.

The investigation described below was undertaken in order to obtain a more accurate knowledge than can be obtained from any published results as to how the resisting torque due to the blades of a fan dynamometer varies with their shape and size, with the diameter of the fan, and with the density of the fluid in which the blades rotate.