_Lax condition of the Shoulder-Joint in Bats, Birds, etc._--The great laxity of the shoulder-joint in bats and birds, readily admits of their bodies falling downwards and forwards during the up stroke. This joint, as has been already stated, admits of movement in every direction, so that the body of the bat or bird is like a compass set upon gimbals, _i.e._ it swings and oscillates, and is equally balanced, whatever the position of the wings. The movements of the shoulder-joint in the bird, bat, and insect are restrained within certain limits by a system of check ligaments and prominences; but in each case the range of motion is very great, the wings being permitted to swing forwards, backwards, upwards, downwards, or at any degree of obliquity. They are also permitted to rotate along their anterior margin, or to twist in the direction of their length to the extent of nearly a quarter of a turn. This great freedom of movement at the shoulder-joint enables the insect, bat, and bird to rotate and balance upon two centres--the one running in the direction of the length of the body, the other at right angles or across the body, _i.e._ in the direction of the length of the wings. In the bird the head of the humerus is convex and somewhat oval (not round), the long axis of the oval being directed from above downwards, _i.e._ from the dorsal towards the ventral aspect of the bird. The humerus can, therefore, _glide up and down_ in the _facettes_ occurring on the articular ends of the coracoid and scapular bones with great facility, much in the same way that the head of the radius glides upon the distal end of the humerus. But the humerus has another motion; it moves _like a hinge from before backwards, and_ vice versâ. The axis of the latter movement is almost at right angles to that of the former. As, however, the shoulder-joint is connected by long ligaments to the body, and can be drawn away from it to the extent of one-eighth of an inch or more, it follows that _a third and twisting movement can be performed_, the twisting admitting of rotation to the extent of something like a quarter of a turn. In raising and extending the wing preparatory to the downward stroke two opposite movements are required, viz. one from before backwards, and another from below upwards. As, however, the axes of these movements are at nearly right angles to each other, a spiral or twisting movement is necessary to run the one into the other--to turn the corner, in fact. From what has been stated it will be evident that the movements of the wing, particularly at the root, are remarkably free, and very varied. A directing and restraining, as well as a propelling force, is therefore necessary.