the layers of burnt wood and earth, it appeared as if the body had been wrapped in turf, then burnt with the sods and brushwood, and the relics deposited on a flat stone of Lynmouth rag, such as is often brought in ballast from the north of Devon. These may have been the ashes of a Viking or Sea rover of an earlier date than the predatory Danes, the ramparts of whose encampment may be seen from this spot on the jagged edge of the cliff near the Nash Point, almost in a line with the higher lighthouse. Another mere or boundary stone formerly stood about 200 yards to the westward, but it has been built into the wall of Pickett Lease.

A third similar mound gives its name to the Bredbarrows, or broad barrows, a field on the south side of Priest Lane, leading from Nottage to the Hutchwns. The tenant, having his horses unemployed, dragged down a neighbouring upright stone, or maen hir, which I remember having seen many years ago. From the vestiges of broken pottery in the fields between this place and the Pickett Lease, near Porthcawl, it is probable that some other mounds have been dug up to dress the fields, mixed with lime, and scattered over the light sandy soil.

A fourth probably existed at the Summer or Samar barrows, to the north of Priest Lane, where there are traces of a ridge or mound, and two or three furlongs on, in the same direction, there is a round low tumulus in the Little Harolds. The gradual levelling and filling in the Great Harolds have left in it no similar vestiges of the past. To this list of sites of ancient mounds, a sixth should probably be added, viz., the Twmpath Ddaear, before mentioned, as almost at the extreme northern boundary of the parish on Newton Down. King's Hill, near South, seems alone to retain a name from the ancient chieftains of Morganwg in Tir y Brenhin. Perhaps from these vestiges it may be fair to infer that Nottage, near to which, at St. David's Well, and also to King's Hill Well, there is an unfailing supply of fresh water, was first permanently inhabited, long before the Newtown in

Nottage, as it is called by the Welsh, was founded by the Normans.

It may be worth while to add that the old Celtic and Scandinavian Beltain, or bonfire, seems not to have been unknown. On my making inquiries about a small circular enclosure to the south-west of Newton, or Sanford Well, I was told about the year 1820, by the old people, that there had been a custom of kindling a fire in it annually on midsummer day, throwing a small cheese or cake across it, and then jumping over the embers. This custom of leaping through the flames as an expiation is reprobated by Theodoret on 2nd of Chronicles, also by the lxv. canon of the Synod of Trulla, or Tulla, and the commentators on it, and by Balsamon and Zonaras. It was as old as the Roman Palilia, or perhaps as the Phoenician rites. Something of a superstitious notion of protecting the crops from blight was attached to this ancient observance.5 Port Eynon, in Gower, is the only place in which I ever witnessed an approach to this rite, where I saw the younger inhabitants lighting fires on the sandhills at midsummer many years ago. I could obtain no other explanation than that it was customary. The stones from the enclosure on Newton Sands have been taken long since to repair the road to the port. The foundations which I observed were obviously distinguishable from those of the raised kilns for burning kelp, which were smaller and stood in rows together nearer the sea; their ruins have now almost disappeared.

ROMAN REMAINS.

We now arrive at another class of ancient remains, not hitherto noticed, or even suspected to exist.

In quarrying the rock and making considerable excavations at Nottage Court, one of my workmen found at the foot of an old wall on the north side of the house, a

4 "Moxque per ardentes stipulæ crepitantis acervos.
Trajicias celeri strenua membra pede."

Ovid, Fast. L. iv. 781. 5 See Pennant's Tour in the Highlands.

small brass head, or rather face of Medusa. Seven vipers' heads form the extremities of the little sconce or mask, and there are two small holes at top and bottom for fastening it. (Fig. 2.) In the forelock, parting from the forehead, and in the tie of the wreathed snakes under the chin, there is a resemblance almost exact to the heads of the Medusa, figured in Mr. Lee's interesting little book on Roman Remains found at Caerleon, 1850, (plate viii. fig. 2, vi. fig. 1, Archæologia Cambrensis, vol. iv. First Series.) Those discovered at Bath are so similar, that there seems little room for judicious scepticism. A ring, which I did not see, is said to have been found in the Court at Nottage before my diggings commenced. A large collection of fragments of shells which we found, might have accumulated during the occupancy of mediæval, or perhaps still earlier, inhabitants. A small rude figure of baked clay (Fig. 3) was found last summer, in a clod of earth brought from the Lock's Common for the vinery. It is in character not unlike the bronze figures of animals represented by Mr. Lee, in plate iv. figs. 1 and 2, and is evidently intended for a sheep.6 This earthen figure may have been an offering for the safety of the flocks, and still earlier than those of bronze. In the illustration it is faithfully delineated by Mr. O. Jewitt. It seems considerably older than the fictile mounted knight found at Lewes in 1846, for a sketch of which, from the Archæological Journal, vol. iv. I am indebted to a kind and intelligent friend.

Near Dan y Graig House, about half a mile from Newton, still more satisfactory discoveries have recently been made. In removing a bank in order to improve the grounds in the year 1850, a coin of a Roman empress, much worn indeed, but distinguishable by the head dress reaching towards the back of the neck, was dug up. (Fig. 4.) The finders did not spare the use of sand and filing to ascertain that it was not gold, though it may have been gilt; it was obligingly given me by Miss

6 Archæologia Cambrensis, vol. iv. First Series, p. 79.

Turberville. Two pieces of stucco from the interior of a room, with signs of a diamond pattern in blue and dark ochre, an iron key about four or five inches in length, some nails with pyramidal heads in a piece of board, and pieces of cement for flooring, strewed with pounded brick or tile, were also found. These remains certainly seemed Roman or Romano-British. Tradition speaks of the site of an old house near the Ridge, under the large elm, where these things were discovered. It was on the left, or north side, of the occupation road which continued from the main road towards the foot of the Graig, and then joined Bistil Lane, long since taken into the fields. The "Rhwsted," or house-stead, was the name of the old barn close at hand; near to this was an isolated sling of land belonging to the Nottage Court property, till exchanged away, previous to the sale of Dan y Graig, in 1839.

Whether some officer from the cohorts quartered in the Roman camp above Pyle, or in the outwork near Heol Sheet on Newton Down, was tempted by the sheltered aspect and pleasant sea view to fix his residence here, or whether some British chief, unmolested whilst he paid taxes to the Roman authorities, resided in this part of the extensive tract called Tir y Brenhin, (the king's land,) it is now unavailing to inquire. The Roman coins seem to have long circulated in this country. Hence we may probably explain why the Welsh princes had no mint of their own. As soon as Roman coins became scarce, the Saxon money would take its place in a tributary nation.

It may be well to record, even if rather out of place, that when the foundations of Dan y Graig House were cut, under the upper soil a vein of drift sand was found, then supposed to have been carried up across the cultivated land for a distance of more than half a mile, and deposited by the strong south-westerly gales. Subsequent observation has suggested that this sand was more probably a vestige of a raised sea margin, or an indication of the sinking of the level of the sea at some very remote

ARCH. CAMB., NEW SERIES, VOL. IV.

period. These raised beaches have been traced in the north-western district of Somersetshire, and still more extensively elsewhere. There is, or was, a local tradition that the sea shall return, and ships be moored to a sycamore tree growing on the top of Newton Clevis. The nearest modern approach to fulfilment of this prophecy" was made at an extraordinary high tide, when the salt water poured into the celebrated Newton Well. It is probable that the Ogmore once entered the sea far to the westward of its present outlet, and much nearer to Newton. The prediction may be one of Twm Evan Prys, of local celebrity. (Iolo MSS. pp. 200, 616.)

In our next chapter we shall endeavour to follow the descent and division of property in these two little hamlets, Nottage and Newton, from the Norman conquest of Glamorgan to the great social changes of the reign of Henry VIII., aided therein by documentary evidence.

WENLOCK PRIORY.

THE very name by which common consent designates the monastic ruins at Wenlock is characteristic of that general uncertainty or misapprehension which exists as to their foundation and history. The religious house now known as Wenlock "Abbey" belonged to a class whose members in England used uniformly and correctly to be styled priories, as being subject to no resident authority other than that of their respective priors. These priories owed common allegiance to the great Burgundian abbey of Clugny; nor was this allegiance

7 A bard who, falling asleep on Margam mountain, or Crug y Diwlith, the Dewless Hillocks, awoke (like Piers the Ploughman on Malvern) with an unperused book of predictions under his head, used to be cited as authority for the rising of the level of the Channel. A sycamore was planted in front of a cottage on the Clevis many years ago, to be ready for the occasion.