Equally interesting, however, for our purposes is the fact that, though the _ac_-termination occurs frequently in the departments between the Garonne and the Pyrenees, no dolmens exist in that region except, as before mentioned, a few at the roots of the mountains. This, at first sight, might seem to militate against the universality of the theory; but I, on the other hand, only take it to express that the _ac_-people were driven from that country by Ibero-Aquitanians before they had adopted the fashion of stone monuments. If we knew when Aquitania was first occupied by the people whom Cæsar and Strabo found there, it would give us a date before which dolmens could hardly have existed; but as we have no materials for the purpose, all that can be said is that, just as the dolmen races were cut in two by the Belgæ before the use of stone for funereal monuments had been introduced, so here the same phenomenon occurred, and the people we have to deal with were driven north of the Garonne, west of the Rhone, and south of the Seine, before they took to building dolmens--assuming, of course, that they once had extended beyond those limits; but this, except in the case of Aquitania proper, does not at present seem capable of being proved. Before the Romans came in contact with them, and our first written accounts describe them, they had ceased to be a nation politically, and their language also was lost, or, at least, except in the one syllable _ac_, we now know nothing of it. If, therefore, it may be argued, the nationality of this people was lost before the Christian era, and their language had become extinct, these monuments must belong to a long anterior period. There are, however, certain considerations which would make us pause before jumping too hastily to this conclusion. There are, throughout the whole dolmen region of the south of France, a series of churches whose style is quite distinct from that of central and northern France. The typical example of this style is the well-known church of St.-Front, Périgueux. But the churches at Cahors, at Souillac, at Moissac, Peaussac, Tremolac, St.-Avit-Sénieur, and many others, are equally characteristic. The cathedral at Angoulême, the abbey church at Fontevrault, and St.-Maurice at Angers,[393] and the church at Loches--all these churches are characterized by possessing domes, and the earlier ones by having pointed arches which look very much more as if they were derived from the horizontal arches of the tumuli than from the radiating arches of the Romans, which the Celts everywhere adopted; and, altogether, the style is so peculiar that no one the least familiar with it can ever mistake it for a Celtic style. All belong to the same group, and as distinctly as, or even more so than, the _ac_-termination, mark out the country as inhabited in the eleventh and twelfth centuries by a people differing from the Celts. Though, therefore, both their nationality and their language may have