A HISTORY OF SUFFOLK scratches of the group being strictly parallel to one another and frequently running for an inch or more across the surface. There may be one group or several on the same stone, the groups not being necessarily parallel to one another. It is difficult to look at these groups of fine parallel scratches, bearing in mind the extreme hardness of the fine flint surfaces on which they have been engraved, without the conviction that they have been made by the dragging of fine sand (quartz grains) across the surface under pressure. 6. The last class into which scratched implements fall is the most numerous of all. It consists of implements, with slight patinas, but generally sub-lustrous, on which there are two or three slight and, in most cases, fine scratches, generally less than i in. long, which suggest to the observer that what- ever may have been the conditions which gave rise to the scratches on all the various classes of stones just enumerated, these conditions have been present in much less degree in this our last class. As the variety and delicacy of make of implements gradually increased, as we advanced from one class to another, so here we reach the highest point in these respects attained by scratched implements. Even the more delicate arrow-heads are represented by a few specimens. The vast majority of arrow-heads belong however to the great class of unpatinated, unscratched, and highly-finished implements. The above are the six great classes into which the scratched flints may be divided. It is quite certain that the classes are not contemporary with one another, but that there is a sequence in their relative ages ; and it is practically certain that the sequence in which they are placed above is the correct one — Class I being the earliest and Class 6 the latest. Proof of this is to be found in a careful study of the large series of doubly patinated pieces where two classes are represented on the same stone. As the result of this definite sequence, the stones composing the earlier classes must have existed during the subsequent periods when the later kinds of scratching were acquired, and therefore ought to present all the varieties of scratching, not only that of their own period, but those of all subsequent periods. As a matter of fact this is very commonly the case ; nevertheless it is by no means rare for an imple- ment to show nothing but the scratching belonging to its own special period — to have come down true, so to speak. At first sight this seems rather surprising. There is, however, a quite simple explanation of the phenomenon ; but as this explanation is closely connected with the general theory of the causation of the scratches it is not possible to go into it here. The question may be asked : Are these scratched implements really neolithic ; or do they represent some late phase of the Palaeolithic Period ; or perhaps even an intermediate period.? The answer is that with the ex- ception of the first class, a few examples of each class have been found with parts of their surfaces polished. Care must of course be taken in such cases not to confuse the scratches due to artificial polishing with the scratches due to some quite outside cause. There is really no difficulty in making the dis- tinction between the two, as there is practically no resemblance between them. In certain instances the ' natural ' scratches, if they may be called so, run right athwart the ' artificial ' scratches ; and the contrast between them becomes at once evident. With regard to the distribution of these scratched flints, as has already been stated, they abound in the little valley, but they equally abound in the 252
