Page:Prehistoric Times.djvu/97

Rh The operations of modern gun-flint makers give us a very clear insight into the mode of manufacture of ancient flint implements, and the process is one of considerable interest.

If we take a rounded hammer, and with it strike on a flat surface of flint, a conoidal fracture is produced, the size of which depends, in a great measure, on the form of the hammer. The surface of fracture is propagated downwards through the flint, in a diverging direction, and thus embraces a cone, the apex of which is at the point struck by the hammer, and which can afterwards be chipped out of the mass. Flint cones, formed in this way, may sometimes be found among heaps of stones broken up to mend the roads, and have doubtless often been mistaken for casts of fossil shells.

If a blow is given, not on a flat surface, but at the angle of a more or less square flint, the fracture is at first semiconoidal or nearly so, but after expanding for a short distance, it becomes flat, and may be propagated through a length of as much as 13 inches, thus forming a blade-like flake (figs. 86-94), with a triangular cross section (fig. 95). The consequence is, that a perfect flint flake will always have a small bulb, or projection (fig. 88, a), at the butt end, on the flat side; this has been called the bulb, or cone, of percussion. After the four original angles of a square block have been thus flaked off, the eight new angles may be treated in a similar manner, and so on. Fig. 86 represents a block, or core, from which flakes have been struck off. A very long flake in my collection, from Fannerup in Jutland, is figured,