Page images

ground before the older cuttings were made is now best represented by that part of the section which stands at right angles to the Ystwyth. It shows a maximum depth of 4 ft. of shaley loamy soil overlying material which is of a somewhat gravelly texture, with a clay matrix. The latter is exposed to a depth of 4 ft., and is probably of glacial origin. The line of demarcation between the upper and lower layers is, however, not conspicuous, and the thickness varies somewhat from point to point. Flint chips have been.


Fig. 2.-Site of a Prehistoric Flint Factory, Aberystwyth

rarely found near the upper surface of the lower layer; but they are common throughout the upper layer, not only in the hospital section, but also in the sides of the cutting leading to the bridge and in the cliff overlooking the Ystwyth.

The majority of the flints and other articles were found on the tip made from the excavation of the site and at the foot of the cliff immediately above the bridge, where they appear to have fallen with the slipping of the soil, and have there been washed by the river at high tides. I have carefully searched the beach at other places, and also the soil in the neigh

bouring gardens between the hospital site and the Great Western Railway line, which is about 60 yards to the east, but have, as yet, been unable to find any flint chips; they are, therefore, confined to a limited area.

The various objects found are roughly classified below according to definitions given in Sir John Evans' Ancient Stone Implements of Great Britain (2nd Ed., 1897), and references are given in square brackets to the figures in that book.

Arrowheads.-Several of the flint flakes may be referred to arrowheads of a rather crude pattern. They exhibit in certain cases fine and careful chipping round the margin, and the best specimens show a high degree of symmetry. They vary in length from in. to 1 in. The majority, by far, are of the lanceolate, or leaf type, a few only being of the broad type [Figs. 193, 288]. Though the best chipped flakes show great fineness of work they may not represent finished products. No barbed arrowheads have been found.

Scrapers. A scraper is defined by Sir John Evans as "a broad flake, the end of which has been chipped to a semi-circular bevelled edge round the margin of the inner face, similar in character to that of a round-nosed turning-chisel." Some of those found on this site are finely chipped along all the edges [Figs. 207, 208, 219, 226 A]; others only along part of the edges [Figs. 206, 213]. It is possible that many of these implements which are here classified as "scrapers" might also have been used as "strike-a-lights." They vary in diameter from 1 in. to 13 in. and are relatively thick.

Flakes. This group includes only those "artificial splinters of flint which, either in their section or outline, or in both, present a certain amount of symmetry and appearance of design." of design." About fifty typical specimens are classified as "flakes," but many of these might have been intended for scrapers. They vary in length from 1 in. to 3 in., and in width from in. to 1 in. [Figs. 399, 400, 413 D]. Unlike the majority of those

classified as arrowheads, they often show a well developed bulb-of-percussion. They are generally two-edged, but may be one-edged only. Either the bulb end or the opposite end is often worked down by secondary chipping to a rather fine edge. The longer edges are sometimes worn, apparently by attrition. The unused flakes occasionally show a remarkably sharp edge. The bulb face is frequently flat or slightly convex, but it never shows signs of secondary chippings; the opposite face, which is more convex than the former, consists generally of numerous secondary concave chippings most of which were probably executed previous to the detachment of the flake from the parent core. Unlike the arrowheads they are seldom finely worked on their long edges; and this, together with the evidence of their edges being worn by attrition, suggests that some were used as scrapers of soft material and others as knives.

Chips and Spalls.—These are "the ruder forms, such as would result from chipping some large object into shape, without any regard to the form of the part removed." Such chips must have been produced in the manufacture of various articles; but it is impossible to say for individual specimens whether they were produced intentionally, or merely represent waste chips. They have no character sufficient to attribute them to definite implements.

Cores.-These vary in height from in. to about 21 in. They serve beautifully to show the manner of detachment of a flake from the parent flint, and point to the dexterity that was essential in order to excel in the art of making flint implements. A typical flint core nearly always has a flat, and more or less circular base, from which it tends to taper to a blunt point forming a pyramid. Many of those found compare favourably with Fig. 189 (Evans). aboutin. high; and it is evident that a number of flakes of small size have been chipped off this specimen, indicating that many of the flakes which are

The smallest is

referred to arrowheads were made intentionally, and are not merely accidental products.

Various other flint articles have been found which cannot be readily referred to any of the above classes. Other articles discovered include various well-shaped pieces of chert, worked in the same manner as flint; vein quartz, apparently worked; a sandstone, worked at the edges; pieces of pottery, and fragments of bones; miscellaneous articles, such as lumps of lead, covered with a thick white coat, one piece weighing about 24 ounces; a bronze buckle, about 1 in. long, showing file markings probably medieval; numerous pipestems, etc. It is not suggested that all the articles

found on the site are coeval.

With regard to the age of this site, a comparison with similar flint factories recorded from Pembrokeshire' and elsewhere suggests that it dates from the Neolithic period, but the presence of the other articles shows that the site was occupied for some purposes in much later times.

The above list is by no means complete, but it is proposed to publish a full account at an early date. Having regard, however, to the rarity of flint finds in Cardiganshire, it is thought that this preliminary account may be of some interest, and may stimulate others to look out for similar sites, particularly near the coast, where a supply of raw material was available in the beaches.

The majority of the articles found are exhibited in the Museum of the University College of Wales, and it would be useful if subsequent finders of worked flints at or near the above site would kindly bring them to the notice of the Curator of the Museum, so that all possible evidence of prehistoric man in this locality may be recorded.

1 Law's Little England beyond Wales, Chapter II.



THE Roman road from Deva (Chester) via Varae of Antonine's Itinerary to Kanovium on the River Conwy, and thence to Segontium, has not hitherto, with the exception of the portion near to Chester, been systematically worked out.


In various maps, drawn to show ascertained roads in Britain, different courses are given of that one leading from Deva into North Wales. Mr. T. Codrington (Roman Roads in Great Britain, pp. 87, 88) cites Mr. Shrubsole's view' that the line lay vid Hawarden to Flint, and "thence by a doubtful course Caerwys, where he places Varae. His suggestion commands consideration, for the route has been carefully worked out to fit in with the length of journey set forth in the Itinerary of Antoninus, but the lines are very devious. He also assumes that Caerwys may be converted into Varae, a matter only for etymologists to determine. The Rev. George Herbert (p. 392) suggested a different way, and onwards to St. Asaph, proceeding thence vid Bettws yn Rhos and Gofer, on high land to the Conwy River. But, assuming that the Hawarden-Flint route was for commercial purposes only, i.e., minerals, and the Romans assuredly made many roads, an alternative road vid Mold-to Denbigh, and continuing via Llansannan and Llangerniew to the Conwy seems the most direct course for military service, while it has the advantage of an easier alignment, and if in the past bog-land intervened, as Mr. Shrubsole avers, in opposing such a route, that character of obstruction Roman engineers were capable of dealing with, and did


Arch. Camb., 1892, p. 257.


« PreviousContinue »