to the shaft by several mechanical processes. Sometimes the proximal end had a wedge-shaped slit into which the shaft penetrated an enlarged example of the typical Aurignacien point (PI. V., Nos. 1-7). This process was sometimes reversed by making the slit in the shaft. In other cases the attached end of the lance-head terminated in a long, slanting splay, so as to be spliced with a corresponding one in the shaft (PI. VIII., Nos. 13, 14). Others had the proximal end terminating in a blunt cone (Nos. 1-4), so as to form a loose joint in which case the blunt end of the head had either a circular ridge, or two projecting lobes, or a small perforation placed a little above the cone, which served for the attachment of a string when the head was intended to remain in the hunted animal's body. When spearing salmon the string was attached to a float which indicated the position of the fish after death. These arrangements were perhaps more applicable to the barbed harpoons, of which so many beautiful specimens were found by Lartet and Christy on the Magdalénien stations of the Dordogne. Certain cross striae, often to be seen on the slanting proximal ends of the earlier lance-points, are supposed to be owners' marks {marques de chasse). Some of these barbed harpoon and dart-points are so small as to be sometimes described as arrow-points ; but there is no evidence that the bow was known to the Palæolithic people. Besides, if arrow-heads of flint had been used there is no reason why they should have been abandoned for those made of bone or horn. That the former were more suitable as arrow-tips than the latter is proved by their survival into historic times. The bow, if it had been used in Palæolithic times, would have been made of wood, and few, if any, objects of that material have survived the process of natural decomposition. No spear-shafts, handles, dishes, clubs, mallets, or worked timbers of any kind have been recorded among the relics of Palæolithic man ; but yet wood must have been largely utilised for all these purposes. Among the objects illustrated on cave walls are wooden erections of some kind. (See Fig. 84.)
Both spears and harpoons were thrown by the hand, but this action was sometimes assisted by an apparatus called
