On the other hand, we have been entirely mistaken in reference to the mechanism which lowers the wings. The muscle concerned, that is _kd_, is not at all the antagonist of the elevator muscle of the wing, since it is placed close by this latter, but nearer to the thoracic wall. But then, how does it come to be the counterpart of its neighbor? In fact, the lever of the wing is situated in the projecting piece alone. The extensor muscle of the wing does not pull on the power arm, but on the resistant arm on the other side of the fulcrum (_c_). The illustration shows, however, how such a case is possible. The membrane of the joint fastening the wing-stalk to the thorax is turned up outwards below the stalk like a pouch. The tendon of the flexor of the wing passes through this pouch to its point of attachment (_c_) lying on the other side of the fulcrum (_d_). Thus it is very simply explained how two muscles which act in the same direction can nevertheless have an entirely contrary working power. This is in a way the bare physical scheme of the flying machine by the help of which we shall more easily become acquainted with its further details. Dragon-flies are unquestionably the most suitable objects for the study of the muscles pulling directly on the wing itself. If the lateral thoracic wall (Fig. 173) be removed or the thorax opened lengthwise there appears a whole storehouse of muscular cords which are spread out in an oblique direction between the base of the wing and the side of the thoracic plate. There is first to be ascertained, by the experiment of pulling the individual muscles in the line with a pincers, which ones serve for the lifting and which for the lowering of the wings. In dragon-flies the muscles are arranged in two rows and in such a way that the flexors or depressors (_s_, 1 bis) cling directly to the thoracic wall (compare also the muscle _dk_ in Fig. 172 and _se_ in Fig. 174), while the raiser or extensor (_h_ 1, to _h_ 2, Fig. 172, _hi_ and Fig. 174 _he_) lie farther in. The form of the wing-muscles is sometimes cylindrical, sometimes like a prism, or even ribbon-like. However, the contracted bundles of fibres do not come directly upon the joint-process we have described, but pass over often indeed at a very considerable distance from them, into peculiar chitinous tendons. These have the form of a cap-like plate, often serrate on the edge, which is prolonged into a thread, which should be considered as the direct continuation of the base of the wings. The wings, therefore, sink down into the thoracic cavity as if they were a row of cords ending in handles where the strain of the muscles is applied. [Illustration: