The method of manufacture[76] is as follows:—A block of flint is broken by means of the quartering hammer in such a manner as to detach masses, the newly-fractured surfaces of which are as nearly as possible plane and even. One of these blocks is then held in the left hand, so that the edge rests on a leathern pad tied on the thigh of the seated workman, the surface to be struck inclining at an angle of about 45°. A splinter is then detached from the margin by means of the flaking hammer. If the flint is of good quality, this splinter may be three or four inches in length, the line of fracture being approximately parallel to the exterior of the flint. There is, of course, the usual bulb of percussion, or rounded protuberance at the end,[77] where the blow is given, and a corresponding depression is left in the mass of flint. Another splinter is next detached, by a blow given at a distance of about an inch on one side of the spot where the first blow fell, and then others at similar distances, until some portion of the block assumes a more or less regular polygonal outline. As the splinters which are first detached usually show a portion of the natural crust of the flint upon them, they are commonly |20| thrown away as useless. The second and succeeding rows of flakes are those adapted for gun-flints. To obtain these, the blows of the flaking hammer are administered midway between two of the projecting angles of the polygon, and almost immediately behind the spots where the blows dislodging the previous row of flakes or splinters were administered, though a little to one side. They fall at such a distance from the outer surface as is necessary for the thickness of a gun-flint. By this means a succession of flakes is produced, the section of which is that of an obtuse triangle with the apex removed, inasmuch as for gun-flints, flakes are required with the face and back parallel, and not with a projecting ridge running along the back. [Illustration: Fig. 2.—Flint-core with flakes replaced upon it.] Fig. 2, representing a block from which a number of flakes adapted for gun-flints have been detached and subsequently returned to their original positions around the central core or nucleus, will give a good idea of the manner in which flake after flake is struck off. Mr. Spurrell and Mr. Worthington Smith have succeeded in building up flakes of Palæolithic date into the original blocks from which they were struck. The former has also replaced ancient Egyptian flakes,[78] the one upon the other. Mr. F. Archer has likewise restored a block of flint from Neolithic flakes[79] found near Dundrum Bay, county Down.