As to the causation of the 'radiation' symptoms, it is difficult to discriminate the effects of neighbouring parenchymatous haemorrhages from those of local vibratory concussion of the nervous tissue. The local character of the signs seems, however, to point to causation by molecular disturbance, resulting from the conduction of forcible mechanical vibration to the brain tissue rather than to upset in the intra-cranial pressure. Again the limited nature of the paralysis observed, sharply defines it from the general loss of power accompanying ordinary cases of concussion of the brain. The similarity of the phenomena to those described in other parts of the body under the heading of 'local shock' is sufficiently obvious. The following instance well exemplifies the condition in question: (55) Wounded at Spion Kop. A scalp wound 3 inches in length crossed the left parietal bone nearly transversely, starting 1-1/2 and ending 2 inches from the median line: the centre of the wound corresponded with the position of the fissure of Rolando. The patient was struck at a distance of fifty yards while kneeling; he fell and remained unconscious an hour and a half. Right hemiplegia without aphasia followed. The wound was cleansed and sutured, and in three days both arm and leg could be moved, after which time the man improved rapidly. Three weeks later when I saw him at Wynberg there was still comparative weakness of the right side, but beyond some neuralgia of the scalp, the man considered himself well. No groove could be detected on the bone on palpation. (This case offers a good example of the ease with which bone injury may be overlooked. The man came over to England 'well;' but while on furlough, two pieces of bone came away spontaneously. He is now again on active service.) _Compression._--Equally rare was it for pure symptoms of compression to be exhibited. This depended on two circumstances: first, the rarity of injuries giving rise to meningeal haemorrhage; secondly, the fact that in nearly every case a more or less extensive destructive lesion was present, at the margins of which less completely destroyed tissue remained, capable of giving rise to symptoms of irritation. Again, as we have seen, free haemorrhage into, or from the walls of, the cavities produced in the brain was not a marked feature, and beyond this the large defect in the cranial parietes was calculated to render a high degree of compression impossible. As the most serious head injuries presented a remarkable similarity in their symptoms, I will shortly summarise their common features. Every degree of mental stupor up to complete unconsciousness was met with, but in some instances where the pulse, respiration, and general bodily condition pointed to speedy dissolution, the patients answered rationally often between moans or cries indicative of pain.