The work to be done by a fan consists in putting a weight--that of the air--in motion. The resistances incurred are due to the inertia of the air and various frictional influences; the nature and amount of these last vary with the construction of the fan. As the air enters at the center of the fan and escapes at the circumference, it will be seen that its motion is changed while in the fan through a right angle. It may also be taken for granted that within certain limits the air has no motion in a radial direction when it first comes in contact with a fan blade. It is well understood that, unless power is to be wasted, motion should be gradually imparted to any body to be moved. Consequently, the shape of the blades ought to be such as will impart motion at first slowly and afterward in a rapidly increasing ratio to the air. It is also clear that the change of motion should be effected as gradually as possible. Fig. 1 shows how a fan should not be constructed; Fig. 2 will serve to give an idea of how it should be made. [Illustration: FIG. 2] In Fig. 1 it will be seen that the air, as indicated by the bent arrows, is violently deflected on entering the fan. In Fig. 2 it will be seen that it follows gentle curves, and so is put gradually in motion. The curved form of the blades shown in Fig. 2 does not appear to add much to the efficiency of a fan; but it adds something and keeps down noise. The idea is that the fan blades when of this form push the air radially from the center to the circumference. The fact is, however, that the air flies outward under the influence of centrifugal force, and always tends to move at a tangent to the fan blades, as in Fig. 3, where the circle is the path of the tips of the fan blades, and the arrow is a tangent to that path; and to impart this notion a radial blade, as at C, is perhaps as good as any other, as far as efficiency is concerned. Concerning the shape to be imparted to the blades, looked at back or front, opinions widely differ; but it is certain that if a fan is to be silent the blades must be narrower at the tips than at the center. Various forms are adopted by different makers, the straight side and the curved sides, as shown in Fig. 4, being most commonly used. The proportions as regards length to breadth are also varied continually. In fact, no two makers of fans use the same shapes. [Illustration: FIG. 3]