This page has been validated.
Chap. VIII.
GROTTES DES FÉES.
341

Birra's grave near Monasterboice, a second from the same neighbourhood, at Greenmount (woodcut No. 81), and they exist in Scandinavia, but their home is Drenthe and the neighbouring corner of Germany. As already mentioned, upwards of fifty examples exist in that province. They are much ruder, it must be confessed, than those of France; but this may arise from the nature of the only material available; they have also the peculiarity of having the entrance always at the side instead of at the end.

So far as their distribution in France has yet been ascertained, the Grottes des fées exist only on the Loire, and to the north of it, in fact in the most northern division of the French dolmen region; while, on the other hand, as they are principally found in Drenthe, or at the southern extremity of the German dolmen field, we may assume that there is some connection between the two, or that there would have been if it had not been severed by the Belgians before those in either region were erected.

One of the finest of the French examples of this class of monuments is that near Saumur, at Bagneux. The walls are composed of only four stones on one side and three on the other, yet it measures 57 feet 6 inches by 14 feet 4 inches across. Another, near Essé, is even larger, though not so regular in plan, nor so grand in the character of the stones. It measures, however, 61 feet by 12 feet at the entrance, increasing to 14 feet over all at the inner end. There is a third at Mettray, near Tours, which, though very much smaller, is curiously characteristic of the form. The immense mass in the centre (woodcut No. 125) and the two smaller which form the roof almost take from it the character of rude-stone architecture. There is a fourth, of a less megalithic character, at Locmariaker,[1] and several others are dispersed over Brittany. It is not possible to know whether the intention may not have been that these, like all smaller chambers, should have been buried in tumuli. These just quoted, however, certainly never were so, but this may have arisen from their having been left unfinished. That at Bagneux, however, could hardly have supported a heavy mass without falling in, and that


  1. All these are represented in Gailhabaud's 'Architecture ancienne et moderne,' ii. plates 7 and 8.