re-enforced from nearly every side, while Teutonic elements have been gradually eliminated.
Another and perhaps even more potent explanation for this localization of the Alpine type in Burgundy also lies at hand. This fertile plain is the last rallying point of a people repressed both from the north and the south. The general rule, as Canon Taylor puts it, is that the "hills contain the ethnological sweepings of the plains." This holds good only until such time as the hills themselves become saturated with population, if I may mix figures of speech. Applying this principle to the present case, it appears as if the original Alpine stock in Burgundy had been encroached upon from two sides. The Teutons have overflowed from the north; the Mediterranean stock has pressed up the Rhone Valley. Before these two the broad-headed Alpine type has, as usual, yielded step by step, until at last it has become resistant, not by reason of any geographical isolation or advantage, but merely because of its density and mass. It has been squeezed into a compact body of broad-headedness, and has persisted in that form to the present time. It has rested here, because no further refuge existed. It is dammed up in just the same way that the restless American borderers have at last settled in force in Kansas. Being in the main discouraged from further westward movement, they have at last taken root. In this way a primitive population may conceivably preserve its ethnic purity, entirely apart from geographical areas of isolation as such.
What is the meaning of this remarkable differentiation of population? Why should the Alpine racial type be so hard favored in respect of its habitat? Is it because prosperity tends to make the head narrow; or, in other words, because the physical environment exerts a direct influence upon the shape of the cranium? Were the people of France once completely homogeneous until differentiated by outward circumstances? There is absolutely no proof of it. Nevertheless, the coincidence remains to be explained. It holds good in every part of Europe that we have examined—in Switzerland, the Tyrol, the Black Forest—and now here in great detail for all France. Two theories offer a possible and competent explanation for it all. One is geographical, the other social.
The first theory accounting for the sharp differences of population between the favorable and unpropitious sections of Europe is that the population in the uplands, in the nooks and corners, represents an older race, which has been eroded by the modern immigration of a new people. In other words, the Alpine Celts once occupied the land much more exclusively; they were the primitive possessors of the soil. From the north have come the Teu-
