87. The projection of the great trochanter is almost entirely removed. That protuberance, being directed upward and backward, is approximated to the spine of the ilium. But if it be pushed in the opposite direction, it readily yields, and then, returning to its proper level, allows the patient to move the thigh. 88. The knee is a little bent. A severe pain always accompanies the motions of abduction, when they are communicated to the limb. If, while the hand is applied to the great trochanter, the limb be made to rotate on its axis, this bony protuberance is perceived to turn on itself as on a pivot, instead of describing, as it does in its natural state, the arch of a circle, of which the neck of the os femoris is the radius. This sign, which was first observed by Desault, is very perceptible, when the fracture is at the root of the neck, less, when it is in the middle, and very little, when it exists towards the head of the bone; these are circumstances, the cause of which it is unnecessary to unfold. In rotatory motions, the lower fragment, rubbing against the upper one, produces a distinct crepitation, a phenomenon which does not however always occur. 89. The point of the foot is usually turned outwards; a position which Sabatier, Bruninghausen, and most other practitioners regard as a necessary effect of the fracture, although Ambrose Pare and Petit have borne witness that it does not always exist. Two cases, reported on this subject by celebrated surgeons, have been thought unfounded by Louis, who has attributed them either to an error in language, or a mistake of the transcriber. But the practice of Desault has fully confirmed their possibility. The first patient whom he had under his care, at the hospital of Charity, after he was appointed surgeon in chief, laboured under a fracture which presented this phenomenon. Many other examples occurred to him afterwards, and he believed it might be laid down as an established principle, that, in fractures of the neck of the os femoris, the direction of the foot outwards is to that inwards as 8 to 2. 90. The common opinion is, that this direction outwards is to be attributed to the muscles that perform rotation. But, were that the case, 1st, it is evident that it would always exist: 2dly, all the muscles running from the pelvis towards the trochanter, except the quadratus, are in a state of relaxation, in consequence of the approximation of the os femoris to their points of insertion: 3dly, muscles in a state of contraction would not allow the point of the foot to be drawn so easily inwards. Is it not more probable, that the weight of the part draws it in the direction in which it is usually found.