What, then, is this group of monuments? Sir Gardner assumes that it is a cemetery of the ancient Britons; but, if so, why are not other cemeteries found in the fertile valleys and plains in South Wales? Why did they choose one of the barest and bleakest hillsides, and one farthest removed from their habitations as a place in which to bury their dead? Why did they not, like the inhabitants of Salisbury Plain, disperse their graves pretty equally over an area of 30 miles by 10? Why crowd them into less than half-a-mile? Without reverting to my previous suggestion of a battle-field, I do not see how these questions can be answered; and if so, I do not think we have far to go to look for its name? As hinted above, Arthur's eighth battle must have been fought in Wales. The name of the place is written Guin (Gwyn), Guinon, Guinnon, Gunnion,[198] which certainly is Welsh; and when we find it immediately preceding the battle of Caerleon on the Usk, and the principal monument still bearing Arthur's name, we may fairly, I think, adopt the suggestion till, at least, a better is offered. Be this as it may, I think all antiquaries will agree with Sir Gardner Wilkinson in assuming that this is the stone of Cetti[199] mentioned in the Welsh triads. 'The 84th Triad' speaks of the Cor of Emmrys in Caer Caradawg (another name for Salisbury), and the 88th of the three mighty achievements of the Isle of Britain, the raising of the stone of Cetti, the building of the work of Emmrys, and the heaping of the pile of Cyvragnon.[200] The work of Emmrys (Ambrosius) is generally admitted to be Stonehenge. If this is the stone of Cetti, which I see no reason for doubting, it only remains to identify the third. Most antiquaries suggest Silbury Hill; and, if I am correct in placing these three monuments so near one another in date, this seems also extremely probable, and so far as it goes, is a satisfactory confirmation of what has been advanced above from other sources. From my ignorance of the Welsh language I am not in a position to say what amount of reliance should be placed in the evidence of these triads. But Herbert and other competent scholars consider it undoubted that Emmrys is Ambrosius, and the 'Work' referred to certainly Stonehenge. If this is so, it fixes its date beyond question, and as the other two are mentioned in the same breath it is probable they were not distant in date. All this may be, I believe certainly is so, but the circumstantial evidence adduced above seems to me so much clearer and so much more to be relied upon, that it derives very little additional force from the utterance of the Welsh bards. It is, however, no doubt satisfactory that their evidence coincides with everything that has been brought forward above, as bearing directly or indirectly on their age or use. * * * * *