not be pushed too far. Thus Ranke of Munich, most eminent authority, has striven for years to account for the broad-headedness of the Bavarian population by making it a product of the elevated and often mountainous character of the country. This being proved, it would follow that the Bavarians still were ethnically Teutonic, merely fallen from dolichocephalic grace by reason of change of outward circumstances. This theory seems to be completely incapable of proof; for, as Ranke himself has shown,[1] the effect of the malnutrition generally incident to an abode at considerable altitudes is entirely in the opposite direction. Among poorly nourished children in factory towns, for example, the immediate effect is to cause an arrest of development about the temples, exactly where the broad-headed Alpine race is so well formed. It is strange to us in America to find how important such matters may become by reason of a social differentiation between races. Another potent example is offered in Russia. The late Professor Zograf, of Moscow, than whom none stood higher as an anthropologist in Russia, confronted by the same division of ethnic types as Germany contains, has positively identified the blond long-headed one as the original Slav.[2] This may or may not be true; it may be gratifying to have it so. To us the evidence apparently points the other way. In Russia, however, no other conclusion than this would be tolerated for an instant. Pan-Slavism prevails even in science.
After this excursus, let us come back to statistics and examine the evidence from the study of blondes and brunettes among the school children. Our double-page map, as will be observed, includes not only the German Empire but Switzerland, Belgium, and Austria, down to the Adriatic as well—exclusive, however, of Hungary. Virchow's great census in Germany was extended over the other countries in quick succession.[3] The system employed was identical in all, save in Belgium; and even here the definition of brunettes was the same, although the term blond was made more comprehensive. For this reason the results are strictly comparable so far as our map is concerned. A great defect in all such investigations on children, as we have already stated, lies in the tendency to a darkening of hair and eyes with growth. This is probably intensified in the more southern countries, so that our shading probably fails to indicate the full extent of the progressive brunetteness in this direction. North of the Alps, however, we may accept its evidence provisionally, at all events.
