However this may be, the mongoloid theory can no longer be accepted as absolute. The man of Cro-Magnon and that of Furfooz cannot be placed in the same group, and considered as belonging to the same race. The idea of M. Pruner Bey is, nevertheless, partly true; and the honour of having connected living with fossil populations cannot be denied to this eminent anthropologist. Still, what he has said of one race must be applied to the rest. The inhabitants of Western Europe are connected with the quaternary period, not by a single root, but by six at least, and perhaps more. IX. A methodical distribution of the different races of a species is never an easy task. The difficulty is very strongly felt in the study of living human races; it is still greater in dealing with fossil races. Even if the materials were as abundant as they are rare, we no longer have the perfect individual, and cannot attempt to apply the _natural method_; we are forced to be content with a _systematic classification_. This is what M. Hamy and I have been obliged to do; and without sharing the absolute opinions which were once advanced by Retzius, we took the general form of the skull as the starting point for our classification. In so doing we have, moreover, only imitated palæontologists in their studies upon fossil animals. We have already seen that considerations drawn from this method lead to a division of fossil man into two groups, the one dolichocephalic, and the other brachycephalic. The Lagoa Santa skull, which must from all appearance be the type of a distinct race, is evidently connected with the former. The accounts of this fossil are, however, at present so incomplete, that I cannot stop to consider it in such a rapid sketch as this. In these two fundamental groups differences exist side by side with the common character. In the former these differences are very great and strongly marked; they are generally less so in the latter. Thus we have clearly distinguished the two dolichocephalic types, while we have placed in the same chapter, and as it were in a kind of family, part at least of the brachycephalic races. Several objections may be raised against this nomenclature, of which we are well aware. We understood perfectly that the skull of La Truchère is as distinct from those of Furfooz as the Neanderthal skull is from that of Cro-Magnon. On the one hand, however, this skull is the extreme limit of a graduated series, from which it seemed to us difficult to detach it; on the other, this fossil, at the time when we were writing, was perfectly unique. Even at the present time it has only again been met with in the neolithic period. Thus, in giving it a place in our table, we did not wish to separate in an absolute manner an individual case.