weapons and broken fragments. Information of the discovery was given to Walter D. Jones, Esq., M.D., of Glancych, who immediately went to the spot, and obtained possession of the collection, of which he sent a brief notice to the Archæologia Cambrensis, vol. vii, p. 313. They were subsequently exhibited at Swansea during the meeting of the Association. The exact spot of their finding was Pant-y-maen (the hollow of the stone) near Henfeddau farm, and lies exactly between two small earthworks of military character. There is a local tradition of a battle fought near Pantgwyn (White Hollow), which may, however, have derived its origin from the name of Henfeddau, and frequent discoveries of sepulchral urns at Pantgwy. This important discovery, however, of so many military weapons, between two small earthworks, seems to confirm the truth of the tradition.

The Pant-y-maen (or as they may be called from their present place of keeping, the Glancych) weapons present a striking contrast to the contents of the Guilsfield deposit, and are evidently of a much earlier date. With the exception of one article, which is of copper, they are all of a bright yellow bronze, some of them still retaining that peculiar lustrous lacquer which is often found, especially on sword blades. What this kind of varnish is, has not yet been ascertained, but is thought by Dr. Wilde to have been of a vegetable nature. No traces of ornament exist on any of the weapons, unless some rude indentations on a spear head may have been intended for such. The contrast also between the ferules of this collection and those of the Powis collection, is very remarkable-not only from the great difference of lengths, but in their form. Some of them still retain portions of oaken shafts.

Another difference to be observed is their mutilated state, as compared with the Guilsfield ones. Dr. Jones thinks they have been thus broken in battle, and have been buried along with some fallen chief. From the manner in which they have been bent and fractured (as

especially in the case of the sword) it is more probable that the destruction has been deliberately effected in honour of the dead with whom they were buried. Such a custom existed generally in the earliest times. Thus, where stone implements have been found in chambers of the earliest character, the finest and largest of them have been broken in halves, as was the case in the chambers of Tumiac and Mont St. Michel, in Morbihan. The same custom remained till the time of our Saxon ancestors, and the Merovingian period of France. It is probable that the same custom was observed in burying a chief at Pant-y-maen.

It will be seen from a brief enumeration of the articles found how complete the destruction had been, although to form an adequate opinion of the manner and violence with which they must have been thus treated, it would be necessary to inspect them personally. The swords, in particular, have not only been broken up, but twisted and bent various ways, which could only have been done with great violence, and certainly not in the ordinary fighting of the day.

1.-Leaf-shaped sword in three pieces, having its extreme tip and the lower portion of the handle blade broken off.

2. Three portions of a similar sword, but without any portion of the handle plate.

3.-Upper fragment of another sword.

4.-The handle-plate of a sword, probably belonging to No. 2.

5.-Four ferules, two of them crushed in, and having their open ends broken. These retain their wooden shafts.

6.-Sockets of five spear heads, perfect at the lower, but mutilated at the upper ends; four of them retaining the lower part of the shoulders of the blade.

7.-One spear head, having its tip broken off, but found with it.

8.-One very short lance or spear head perfect, except that the faces of the blade have been battered in.

9.-A similar one, but the lower part of the socket battered and partly broken.

10. Four heads of spears or lances; one of them copper, and two with straight and not curved edges. 11.-The central portion of a large spear head. 12.-Tip of a scabbard; perfect.

13. Three small rings; perfect..

14. Various fragments of sockets of spear or lanceheads.

The following have been selected out of the above list for illustration as the most perfect and remarkable. The first of which is the handle plate (No. 1) of a bronze sword, which was in all probability connected with the second sword, mentioned above. The intervening portion, however, between the handle plate and the lower end of this sword being missing, there are no means of ascertaining whether the conjecture is correct. It, however, evidently belonged to that early class of weapons known as the leaf-shaped form. It has six rivet holes, and fortunately one rivet remaining in its place, remarkable for having beth its ends slightly bevelled. As by these rivets the plates of wood, or bone, or metal, were secured to the handle plate, it may be inferred that the thickness of the completed handle was about half an inch, that being the length of the rivet. This, however, must depend on the assumption that the faces. of the rivets were exactly flush with the surface of the handle; and if this was the case, the completed handle itself must have been thicker near the shoulder of the blade than near the pommel. So little is known, as has been previously observed, as to the manner in which these handle plates were covered, that even the presence of a solitary rivet becomes a matter of interest. We should, however, notice the absence of the burr, or exposed and flattened head of the rivet, which Dr. Wilde thinks indicates that the material riveted must have been metal. According to Dr. Wilde's statement, most of the handle-plates of short broad leaf swords terminate in straight T-like projections, exactly as this one does ;.