bottom of the table at the south. Thus, among two thousand and fifty natives of Tunis in North Africa, true Europeans as we must repeat, Dr. Collignon found that, while blond hair or eyes were noticeable at times, in no single case was a pure blonde with both light hair and eyes to be discovered. Similarly, in Sardinia, less than one per cent of the population was found by Dr. Livi to be of this pure blond type. The interest and significance of this undoubted fact lie in its bearing upon the theory, propounded by Dr. Brinton, that northern Africa was the center of dispersion of the blond invaders of Europe, who introduced a large measure of its culture. The same thesis is upheld in the latest comprehensive work in anthropology: I refer to Keane's Ethnology. We shall return to this theory at a later time. It is sufficient here to notice how completely this blond type vanishes among the populations of the south of Europe and northern Africa to-day. Such blondes do occur. Each one in so dark a general population as here prevails is a host itself in the observer's mind. The true status is revealed only when we consider men by hundreds or even thousands.
Thus far we have been mainly concerned with the pigmentation of the hair and eyes as a result of climatic or other environmental influences. Let us now consider the racial aspect of the question. Is there anything in our map which might lead us to suspect that certain of these gradations of pigmentation are due to purely hereditary causes? In other words, do the long heads and the short heads differ from one another in respect of the color of the hair and eyes, as well as in cephalic index? In the preceding paper we took occasion to point out in a general way the remarkable localization of the round-headed element of the European population in the Alps. The great central highland seemed indeed to constitute a veritable focus of this peculiar physical type. In this way it divided two similar centers of long-headedness—Teutonic in the north, Mediterranean in the south—one from another. This geographical characterization of the broad-headed variety entitled it, in our opinion, to be called the Alpine type, in distinction from the two others above mentioned. It will now be our purpose to inquire whether or not the physical traits of pigmentation stand in any definite and permanent relation to the three types of head form we have thus separated from one another in the geographical sense.
Many peculiarities in our color map point to the persistence of racial differences despite considerable similarity of environment. Thus the Walloons in the southeastern half of Belgium, with a strip of population down along the Franco-German frontier, are certainly darker than the people all about. Among these Wal-
