No systematic attempt has yet been made to construct an ethnographic map of the country from the architectural remains, though it is easy to see how it might be done. What is wanted is that some competent archæologist should do for the ethnography of France what Sir W. Smith did at the end of the last century for the geology of England. Like that early pioneer of exact knowledge in his peculiar department, he must be content to wander from province to province, from village to village, visiting every church, and examining every architectural remain, comparing one with another, tracing their affinities, and finally classifying and mapping the whole. It is probable that the labour of one man would hardly suffice for this purpose. Monographs would be required to complete the task, but it is one of such singular interest that it is hoped it may soon be undertaken. One of the great difficulties in attempting anything of the sort at present is the nomenclature. When the science is further advanced, such names as Silurian, Cambrian, &c., will no doubt be invented, but at present we must be content with the political name which seems most nearly to express the ethnographical distribution; though in scarcely a single instance will these be found strictly correct, all in consequence being open to adverse criticism. In France it frequently happened that two or more ethnographic provinces were united under one sceptre— eventually all were merged into one—and during the various changes that took place in the Middle Ages, it was only by accident that the political boundary exactly agreed for any great length of time with the ethnographical. In Germany, on the contrary, a single race is and was cut up into numerous political divisions, so that it becomes, from the opposite cause alone, equally difficult to apply a nomenclature which shall correctly represent the facts of the case. In such a work as this it would be manifestly absurd to attempt to adjust all this with anything like minute accuracy, but the principal features are so easily recognised that no great confusion can arise in the application of such names as are usually employed, and it is to be hoped that before long a better system of nomenclature will be invented and applied.