Immagini della pagina


covered over, can still be traced”. Mr. Wynn Williams, in his account of that small but very remarkable specimen called Pentyrch, Carnarvonshire, informs us that “in the thickness of the wall are traces of at least three chambers,-one apparently rectangular, measuring 6 ft. by 4 ft. ; and the others probably circular.”

At the points B in the plan are the much reduced remains of two short cross-walls' connecting an outer with a corresponding inner rampart, intended either to check the progress of an enemy who had gained a footing between the walls, or possibly as causeways or passages for the advance or retreat of the defenders from one bulwark to the other. They are 8 or 9 ft. wide; and if we suppose their tops to have been covered over with sods or clay, might have been thus used by the active and lightly armed natives. At Dolbury, in Somersetshire (a magnificent camp), I observed a more perfect example of these connecting structures.

The ground-plan at c represents a curious specimen of a partitioned hut, having two compartments of a singular form.

It is situated on the south-eastern face of the hill, immediately above the second rampart, of which irregular work it almost forms a part. The outer wall of this hut deserves consideration, because where it faces the south and west it is of twofold construction, consisting of two walls separately but contiguously built; the outer one touching and embracing the inner one, as if added for support or strength. The outer fold of masonry in its present state is 3 ft. high ; the inner one exceeding it by 2 ft., and forming a kind of upper step. The breadth of the inner wall at top is 4 ft.; and of the outer one, where thickest, about 3} ft.; their combined measurements being 7] ft. As the wall curves round from west to north it disappears under fallen stones; and where it again emerges, on the northeastern side, it is single. It was not without interest, I must confess, I observed even this small specimen of what may be called the Firbolgic style of building within the limits of Braich y Ddinas,-a town so dis




tant from Somersetshire and the South Isles of Aran, where double and triple walls are met with. Slight as the indication is, it still upholds the opinion of Professor Babington, long since expressed in the fourth volume of the Archæologia Cambrensis, Third Series, where he writes, “a careful comparison of the Firbolgic forts of Aran and Dingle with the ruins at Tre 'r Ceiri, Penmaenmawr, Carn Goch, etc., will probably result in a conviction that they are works of the same race”. And further on : “I have myself examined the stone forts and towns in the counties of Caernarvon and Caermarthen, and think, as has been already observed, that antiquaries who have had similar opportunities will be unable to avoid the conclusion that they were raised by the same or a closely kindred race with that which built the stupendous Irish duus.”

It may not be generally known that scarcely four miles from Tre 'r Ceiri there exists, on a small scale, a thorough specimen of a twofold rampart, resembling in construction and character the great inner fortification at Worlebury so often noticed in this memoir.

On a harvest day, having descended from the ever interesting ruins on Yr Eifl (better known as The Rivals), I extended my walk to the top of a conical hill called Pen y Gaer, at the eastern end of the parish of Llanaelhaiarn, where, instead of the ordinary earthworks I looked for, I found the remains of a small fortified town, defended on its western and weaker side by a double wall with a terrace on the outside, precisely of the same character, but not so imposing in height and construction, as the specimen I had seen at Worlebury. The top of the wall which forms the outer step or terrace of the rampart, is several feet lower than the top of the inner line of stonework; and, like the example at Worlebury, it slopes away remarkably towards its base, serving, whatever its ulterior object, the purpose of a prop or buttress to the inner work. In this respect it is a complete copy of the Somersetshire model. Whether the terrace was designedly made as it now


appears, to be occupied by the first line of defenders, whilst the higher and more commanding interior wall was manned by a second row of combatants, or whether its present appearance is the result of dilapidation, I cannot say ; but it is unquestionably curious that ramparts so far removed from each other should, after the lapse of centuries, retain a form so similar and so pecu·liar.

Another striking resemblance remains to be noticed. I find by my notes taken at Pen y Gaer and Worlebury, which I believe to be correct, that the thickness of their walls is precisely the same; the inner fold of masonry at both places measuring 8 ft. across, and the outer one 5 ft. across, their united breadth being 13 ft. We are indebted to the small size of its stones for the little that remains of the Pen y Gaer rampart, all that was suitable of its materials having been worked into a boundary fence which runs up the crest of the hill. On its northern and eastern fronts it was protected by a precipitous decline of stones and rocks, rendered more inaccessible by a parapet of some kind at top, traces of which still appear. The interior of the town is pitted with a number of circular hollows edged round by banks and protruding stones, marking with certainty the sites of dwellings, the walls of which have been removed. It

It may be well to notice that the rampart of the neighbouring fortress of Tre 'r Ceiri differs in some respects from the one I have been describing, possibly, but not necessarily, indicating thereby a difference of age or race. At Tre 'r Ceiri the main thickness of the wall is on the outer side of the rampart, the narrower and lower terrace being within, which is a reversal of the arrangement at Pen y Gaer. It has not, moreover, the Pen y Gaer and Worlebury batter or slope of its outer face, but rises somewhat perpendicularly from its base.

We may here revert to the inquiry, who were the people who built these fortified towns of Carnarvonshire? To meet this question, it is necessary to ascertain, as nearly as we can, when the Cymry entered Wales. Their earliest appearance in its northern counties is usually stated to have been at the commencement of the fifth century, or soon after the departure of the Romans, when, under the leadership of Cunedda or his sons, a colony of the Strathclyde Welsh obtained for themselves a permanent settlement in Gwynedd. This arrival of the Cymry in North Wales is regarded as an event well established, but it is not equally well defined when and from what point the South Welsh entered the Principality. Some would have us believe they are a medley of all the tribes left by the Romans in Britain. Others, with a greater amount of probability, hold the opinion that they are of Belgic extraction. A few of the Belgic tribes, we are told, were “descended from the Teutons and Cimbri", and of the nation generally, it is recorded “ that they were of German origin”. According to Cæsar, “they differed in language, customs, and laws from the Gauls, their neighbours,” just as the Cymry differ from the Gaels of the present day. He further states that “the sea coast of Britain was peopled with the Belgians”, including, we may suppose, Devon and Cornwall. The Saxons, according to their own record, had to contend with the Wealas or Walum at all points between Kent and the Severn--a contest which lasted a long period. What more likely than that large numbers of the Belgæ of the south coast, and especially those of Somersetshire? and Wiltshire, well known to have been occupied by them, and connected with Wales by so many of her traditions, should, under the pressure of Saxon invasion, have moved westward along the Severn into the border counties of the Principality, gradually displacing the inhabitants, or becoming incorporated with them as friends or relatives, their own language prevailing ?

1 According to one of the Triads the Cymry came from the Summer Country, a region supposed to be in the far east. Gwlad yr Haf, or the Summer Country, is still in Wales the name of Somersetshire.


This view, if established, would account for the existing state of many things amongst us which are otherwise inexplicable. Quite as natural is it that their Cumbrian and Strathclyde cousins, straitened by similar reverses, declining to unite with the Saxons, and pressed on by the Angles, Gaels, and Picts, should, by successive migrations, have sought a home in North Wales near to their kinsmen, aiding them in repelling the Mercians, and in subduing or ejecting the Gwyddyls. These northern Cymry were possibly more directly descended from the Cimbri than their South Wales friends, and hence retained longer their ancestral name. Some have regarded the Cimbri as located at too great a distance from this country ever to have reached it. The irruptions of a. people so named into Gaul before Cæsar's time, prove that they were then unsettled, and as the Cymry of Strathclyde had a tradition that their ancestors crossed the German Ocean, there does not appear to be much difficulty in the matter.

If the preceding remarks, briefly stated, are true, it would appear that the Cymry, as a people, had not proceeded far in the occupation of Wales when the Saxons landed in Britain. Not so far, at least, as Carnarvonshire and its western counties. The suggestion, therefore, that a race called Firbolgs were the builders and first occupiers of these towns, who, after a long settlement, were dispossessed by the Gael or the Welsh, has much to recommend it, and when we compare the facts adduced by Professor Babington with the events of our own history, we cannot do less than admit that they are not Cymric, but are what tradition points them out to be, the retreats of a people called Gwyddyls in this country, who appear to have been relatives, if not identical with the Firbolgs. We are still left in uncertainty whether the Cangani or Cangi of Ostorius's time were the supplanters of these Firbolgs or members of the same family. Their religion appears to have been similar, and we do not exactly know whether the Cymry encountered them on first

« IndietroContinua »