In spite of a considerable experience, I never saw a case of perforation of either the head or neck of the thigh bone. I saw numerous tracks emerging at the side of the femoral vessels and entering at the buttock or vice versa, but never one accompanied either by effusion into the hip-joint or impairment of movement. Considering the regularity with which haemarthrosis occurred when the other joints were crossed, and also the nature of the compact tissue of the neck of the femur, which must have ensured some splintering, I do not think I can have overlooked an injury of this nature. No doubt also the escape of the neck of the bone was explained in some of the cases by the fact that the injuries were received while the hip-joint was in a position of flexion, the bullet passing over the neck of the femur. In two cases of extensive comminution of the upper third of the femur that I saw, the fissures stopped short at the inter-trochanteric line anteriorly, but in one of them a large angular fragment was torn out of the posterior surface of the neck. Excepting transverse fracture every form was met with in the shaft, although I saw only two instances of perforation. One has been already alluded to and was situated in the broadening portion of the lower third, the bullet taking an antero-posterior course. The second is seen in plate XVII. Plate XII. shows an instance of extreme comminution of the upper third accompanied by the presence of two typical elongated fragments. The course taken by the bullet was almost directly antero-posterior, and the wounds were of moderate size even in the case of the exit one. This seems to preclude the possibility of the injury having been produced by a ricochet bullet, while the fact of perforation and escape of the bullet in spite of the serious damage suffered by the mantle points to the injury having been produced at a short range of fire. The patient himself owns to being quite unable to give any estimate of the distance. Although no suppuration occurred, this fracture was very slow in consolidating, and the free comminution with consequent inaccurate apposition led to the development of four inches shortening of the limb. The skiagram was taken about six weeks after the occurrence of the injury, a few days after I first saw the patient; I have, however, had the opportunity of seeing a second skiagram taken some four months later. This is of considerable interest, as throwing light on the mode of union of such fractures. The two elongated fragments in the later skiagram are widened to three times their original breadth, and form buttresses on either side of the point of union, while the irregular ends of the shaft are rounded off, and the mass of fine fragments behind is consolidated. Beyond this the second skiagram shows that the upper fragment, apparently intact in the first, was really split longitudinally, and therefore was far less useful as a point of support