fortresses in the Principality. There are hill forts on the tops of mountains, of irregular shape, and with more or less formidable defences,--the labour, evidently, of permanent possessors of the soil. Then there are equally laborious earthworks on lower eminences, placed, apparently, to guard a ford, or a pass in the hills. Now a strict examination of these structures may show points peculiar to certain eras; and this, after all, is the great want in Welsh archæology, as I am not aware of anything to distinguish a fort erected during the Roman contest from one of the twelfth century, or even later. Is there any Welsh masonry of either this century or the next? (Of course, I exclude ecclesiastical works.) Rhuddlan Castle is said to have been built on an earlier foundation, and so is Diganwy, as well as Diserth; but is there anything in any of these erections from which this earlier date can be obtained ?

The Roman structures are evident enough, but the term station is a bad one, and ought to be disused. Tacitus tells us that præsidia and castella were built to defend the Roman conquests, the former being large, the latter small permanent fortresses, and round either of these settlers might gather and a town be formed. The castra, on the other hand, were simply encampments where the troops entrenched themselves each night when on march, and the two, on Trecastle mountain, described by Mr. Rees, are of this character. I should suppose that the large one was made on the advance of the troops and the smaller in their return; and, if General Roy's observations apply to Wales, another should be found ten or twelve miles off, and the object of the expedition would be probably ascertained.

As to the roads,—what constitutes a Roman road? Is it one used, or one constructed by that people ? Is the Watling Street a Roman road? In Wales, where the possibility of a way is often a mere geological necessity, I should doubt whether the existing roads were not contemporary with Agricola, and it is likely enough that some of them may bave been repaired and improved in later times, and have left traces which a diligent and experienced archæologist would detect; but plans, sections and structural details are absolutely necessary before we can come to any positive conclusion.

These things, as you have justly observed, can only be done by many parties joining in the work, and I trust that your invitation will be cordially and generally responded to. It will supply a sure foundation for Welsh history; and, by restraining that imaginative power which paints the white-robed Druid, in his circle of stones, performing wonderful rites and celebrating unheard-of sacrifices, where, in all probability, neither Druid rite nor sacrifice was ever seen, and confining it to realities which may be seen, measured, drawn and compared together, deductions may be formed, more or less certain, according to the perfectness of the remains, and the knowledge of the observer.-I remain, &c.,

John ROBSON. Warrington, August 17, 1854.

ROMAN ROADS IN CARDIGANSHIRE.

To the Editors of the Archeologia Cambrensis. GENTLEMEN,—The following observations, taken mainly from the Ordnance map, may assist inquirers in laying down the Roman road from Llanio (LOVENTIVM), in Cardiganshire, to Penal, in Merionethshire. The Sarn Elen runs from the former place for several miles in a nearly straight line northwards, and is still used as a parish road. Continuing further northwards, we have, near Lledrod, a tenement called Ty-'n-y-clawdd, which may possibly be on the old line of the road. A mile further in the same direction, a road bearing nearly due north runs up a hill, on the summit of which there is an object marked, apparently representing a small quadrangular entrenchment. This is marked Šarn Elen. It does not appear how the road would get down to the level of the Ystwyth, but Crosswood (Trawscoed) may perhaps mark the place where it crossed it. A tolerably direct line of road leads from Crosswood to Sarnau, and thence on to the sixth milestone on the Devil's Bridge road. Here the road, supposing it to have taken this line, must have descended a steep declivity to the Rheidiol. Pen-rhin and Troed-rhin-felen suggest its course. From this point we have a series of lanes in a nearly straight line, crossing the parallel ridges and vales to Talybont. Maesy-Bangor, Llwyn-Iorwerth, Cwrt and Cynullmawr (the three last ancient mansions, and the first a name suggestive of antiquity), are all upon this line. There is also a strong entrenchment close to it, at Lletty Llwyd. From Talybont to Garreg, where the present ford of the Dovey is, the road may have followed the course of the present one, or a more direct and, in those days, probably an easier route, crossing the hills and falling into the turnpike-road at Furnace. On this line, still marked by a series of mountain lanes, we have the farm of Pen-y-sarn, close to which is a barrow and cist, commonly called Bedd Taliesin. Below this is an ancient arched bridge across the Clettwr, which appears to prove that the route in question was formerly much more frequented than at present. I strongly recommend any Cardiganshire antiquary, who has the leisure to do so, to gauge this line of country with the map in his hand.--I remain, &c.,

W. B. J.

To the Editors of the Archæologia Cambrensis. GENTLEMEN,- The sun-dial exhibited by me, in the temporary Museum, at Ruthin, was found in the village of Clocaenog; and the date, although imperfect, is evidently 1604, the second year of James I.

Some years ago there stood in this village a large old building, called “ Ty-mawr," which tradition says was a residence of the Bishops of Bangor, and built for their accommodation during their occasional visits to this part of their diocese.

The remains of this mansion are remembered by several of the present inhabitants of the village, and described by them as having been a very large quadrangular building, enclosing a small courtyard, in which it is probable the dial stood.

The Bishops of Bangor are lords of the manor,—which will account for the existence of the mansion.

The rooms of the house are said to have been wainscotted with panelled oak; which wainscot was removed when the house was taken down by the family of the present proprietor,--the Rev. John Roberts, of Denbigh.

On the site on which the mansion stood there now stands a smithy and cottage.—I remain, &c.,

ROBERT PIERCE. School House, Clocaenog,

Sept. 11, 1854.

MACHYNLLETH.

To the Editors of the Archæologia Cambrensis. GENTLEMEN,--Can any of your readers give me the authority for the existence of a Roman station MAGLONA, or for its identification with Machynlleth? Further, has any satisfactory etymology been proposed for the latter name? The first syllable suggests the Gaelic Magh, equivalent to the Welsh Maes, or (according to the original orthography) Mages,—the Magus of Romano-Celtic names. I remain, &c.,

YMOFYNWR.

TOMEN Y RHODWYDD. To the Editors of the Archæologia Cambrensis. GENTLEMEN,- In the Brut, opposite to the year 1148, we read:“ Owain ap Gruffydd ap Kynan King of Gwynedd erected a castle in Yale.” This, in the margin of Powel's History of Cambria, is explained as “Castell y Rhodwydd.” And Carnhuanawc, in his Hanes Cymru (p. 619), says that it is now called “Tomen

у

Rhodwydd,” and adds, “no portion of the structure now remains; the only vestiges are the fosses, the ramparts, and the mounds, or earthern hillocks, on which stood the towers." - I remain, &c.,

J. W. AB ITHEL.

Miscellanrois Jotices.

CAREW CASTLE.—As the circumstance of the arms of Philip and Mary being found in this castle appears to have escaped the observation of previous visitors, it is important to notice it, as it is well known that the date of ancient structures and monuments is not unfrequently determined by such apparent trifles. In the present instance it unquestionably affords a clue to the period of the extensive alterations of the windows throughout the interior of the castle, usually attributed to the time of Henry VII., fixing it, moreover, within unusually narrow limits, as the reign of Mary extended only from 1553 to 1558 It is impossible to doubt that these arms and windows are of the same period, the table in which the former are found forming a part of the window over the grand entrance to the old banqueting hall, prior to the erection of the splendid suite of apartments which overlook the estuary, on the north side of the castle. There are three shields in the same entablature, all so much worn as to render them exceedingly difficult to be deciphered. The centre one, even with the aid of a good glass and a bright sunshine, I found impossible to make out, beyond the fact that it was a shield, with supporters a good deal resembling the Tudor griffins, a circumstance which led me to suppose that it was intended for the arms of one of the sovereigns of the house of Tudor, though, had it not been for the side shields, it would have been impossible to have determined the point. Fortunately these were in a less dilapidated condition, which enabled me, after an attentive examination of them, to discover the charges on both; the dexter being France and England quartered in the manner usual subsequently to the Plantagenets, and the sinister, though a good deal more corroded by time and the perishable nature of the stone, exhibiting very unmistakeably the arms of Spain, as borne by Philip II., a circumstance which, beyond all doubt, would fix the date of the alteration to that period. -R. B. P.

WIGMORE CASTLE.--We are sorry to hear from a correspondent that extensive dilapidations are taking place in this fine old castle. Much stone is removing from it to build up walls, &c. We do not know who may be the proprietor of this ancient edifice; but we hope that if this notice meets the eye of any one acquainted with the owner, he will call earnest attention to it. The evil should be stopped immediately.

KIDWELLY CHURCH.—It is with no small satisfaction we have learnt that this valuable church is going to be thoroughly restored, by the mutual consent and aid of all parties concerned in it. A committee has been formed; funds collected; and the management of the

whole matter has been judiciously placed in excellent hands, those of R. Goring Thomas, Esq., of Iscoed. This good example cannot be cited too widely.

CHRIST CHURCH, BRECON.--A rumour has reached us that the old Dominican priory of Christ Church, at Brecon, has been offered for sale by the Ecclesiastical Commissioners; that it has been refused by the parishioners of Llanfaes, whose church is now in ruins; and that it has since been offered to a community not in union with the Established Church. Can any gentleman at Brecon afford us a verification of this rumour ?

Dinas BRAN, VALLE CRUCIS, &c.-A correspondent of the Builder writes as follows, speaking of the condition of Welsh castles and churches :-"Perhaps you will give me space for noticing a cause for their dilapidation, which I think is not generally known, and which cannot be guarded against by ordinary fences, such as are used for their preservation. My remarks apply more especially to the remains in the valley of Llangollen, though they are probably in a great measure applicable to other ruins. I have made many careful surveys of the ruins of Dinas Bran Castle, and I am fully persuaded that the common impression that it is a rudely-constructed building of the local slaty stone is incorrect; and that it has been a carefully-finished structure, with quoins, vaultings, string-courses, and windows of wrought freestone. This stone having been found by the peasantry useful in building ovens and chimneys, for which the local limestone and slate are unfit, they have quarried and mined out almost every fragment, and the result may be easily imagined. When base-tables, and set-offs, and string-courses (which, doubtless, were used of great depth, to serve as bonding to the rubble-work) have been torn out, and the quoins, window-jambs and groining dug out of the walls, and even steps broken short off in the walls, there can be little wonder that the ruin should have a rude and rugged appearance. Though the freestone is now easier to obtain than it was some years back, the work of destruction has by no means ceased; and, unless some means are taken to underpin the walls where they are mined, this picturesque ruin must very shortly cease to exist. Valle Crucis Abbey has not been subjected to the same undermining and sapping process, as the bulk of the work has been freestone, which had only to be picked up from the fallen walls, and carted away; and the small remains of a religious house at Penywern have been saved only because wanted for a barn. The church at Llangollen, which possesses a massive oak roof of Late Perpendicular date, almost covered with rude carved work, is also in a sad state of decay, and has suffered much by "improvements.' A rich screen was cut down about fifteen years ago; and it is a remarkable fact that a very considerable number of the massive oak screens which formerly existed in the churches of North Wales have either been cut down to the level of the pews, or wholly destroyed within the last twenty years.”

a

ARCH. CAMB., NEW SERIES, VOL. V.

2 F