In this example there are two successive decompositions of force. First, the original force of the wind C D is resolved into two, E D and F D; and next the element E D, or its equal D G, is resolved into D I and D H; so that the original force is resolved into three, viz. F D, D I, D H, which, taken together, are mechanically equivalent to it. The part F D is entirely ineffectual; it glides off on the surface of the canvass without producing any effect upon the vessel. The part D I produces _lee-way_, and the part D H impels. [Illustration: _H. Adlard, sc._ _London, Pubd. by Longman & Co._] (86.) If the wind, however, be directly contrary to the course which it is required that the vessel should take, there is no position which can be given to the sails which will impel the vessel. In this case the required course itself is resolved into two, in which the vessel sails alternately, a process which is called _tacking_. Thus, suppose the vessel is required to move from A to E, _fig. 16._, the wind setting from E to A. The motion A B being resolved into two, by being assumed as the diagonal of a parallelogram, the sides A _a_, _a_ B of the parallelogram are successively sailed over, and the vessel by this means arrives at B, instead of moving along the diagonal A B. In the same manner she moves along B _b_, _b_ C, C _c_, _c_ D, D _d_, _d_ E, and arrives at E. She thus sails continually at a sufficient angle with the wind to obtain an impelling force, yet at a sufficiently small angle to make way in her proposed course. The consideration of the effect of the rudder, which we have omitted in the preceding illustration, affords another instance of the resolution of force. We shall not, however, pursue this example further.