Every variety of _fracture of the clavicle_ was met with, even perforation of the most compact portion of the shaft; comminuted, wedge, or notched fractures were, however, the more common, and were accompanied by the development of very large masses of provisional callus during the process of healing. An interesting skiagram is reproduced in plate III., which shows a compound form of injury to the clavicle. The bullet has passed obliquely beneath the acromial end, rising to perforate the posterior compact margin, and producing one of the diamond-shaped openings sometimes occurring in compact bone with the passage of bullets at a low rate of velocity. No case of perforation of the subclavian vein by comminuted fragments of the clavicle came under my notice. _Fractures of the humerus_ of every variety were common, and I think when the statistics of the campaign are published, it will be shown that the humerus was the most frequently injured individual bone in the whole body. I remember to have seen thirteen fractures of the shaft of the humerus in one pavilion alone at Wynberg after the battle of Paardeberg. Perforations of the upper articular extremity were common, and as a rule gave rise to wonderfully little trouble in the shoulder-joint. The outer aspect of the head of the humerus is a common situation for the production of a special form of broken canal or groove (fig. 53). The slope from the greater tuberosity to the shaft naturally favours the production of the injury in this position. I saw only one case in which a vertical fissure extended from a fracture of the shaft into the shoulder-joint; in this case the transverse solution of continuity was at the upper part of the middle third of the bone. Skiagram, plate IV., illustrates a well-marked stellate comminution of the shaft with large fragments. Plate V. shows extreme comminution with fragments blown out of the wound. Two plates, Nos. VI. and VIII., illustrate well the difference resulting from the oblique passage of a bullet at high and low rates of velocity respectively. In both cases good results were obtained; in the more severe the resultant mass of ensheathing callus was very large, temporarily interfered with flexion of the elbow-joint, and consolidation was very slow (see plate VII.). The patient was wounded at Belmont in November 1899, but he was able to row at the end of the summer of 1900, although very prolonged suppuration occurred, and the elbow movements became practically normal. Plate IX. illustrates a transverse track, the bullet having undergone considerable injury during its passage through the bone, as evidenced by the presence of fragments both of mantle and lead in the limb. This might be called an example of transverse fracture, and illustrates the nearest approach to one seen when the bone is struck fairly plumb. [Illustration: PLATE IV. Skiagram by H. CATLING Engraved and Printed by Bale and Danielsson, Ltd. (24) COMMINUTED FRACTURE OF THE HUMERUS