E. TSCHEPOURKOWSKY 
287 
centres isolated by the highest mountains of the globe, from which only the 
northern centre was favourable to the development of brachycephaly. The 
southern remained in a primitive state. 
In the same report and in that read by me before the Anthropological Congress 
in Worms I made the following deductions, based on the measurements of 1,300 
women and their new-born children*: (1) that the head-form of the mother is 
inherited by the new-born child ; and (2) that not only is the form of the skull, 
but also the form of the base determined in this very early period of life. As the 
base is then relatively very small we can hardly speak about its influence on the 
form of the skull : both are equally inherited. 
These last facts brought me more and more to the conclusion that sexual 
selection has played an important role in the variation of the cephalic index. 
Ripley denies this on the ground that it is not a prominent feature of the human 
body. But selection may be at work if the index be simply correlated with 
selected characters. In order to ascertain this, and throw light on many other 
problems which I hope shortly to treat in a special anthropological essay, I have 
collected material for the study of interi-acial correlation. This material is partly 
extracted from Ivanovsky's Anthropological Constituents of the Poj^ulation of 
Russia, partly from Deniker's Races of Man, and as far as regards crania from 
different authors and from my own measurements. From the data given below, 
upon which the cross correlations can also be studied, the correlation tables 
(Appendix, Tables 1 to 28) were formed and the constants calculated, which are 
represented in the following table. Many conclusions can be drawn from this. 
I will mention some of the results. 
1. About the correlation between stature and head form there exist the most 
contradictory opinions between anthropologists. Interracially for the races of 
Russia I found a very sensible negative correlation. The control table for other 
races of the earth was formed on Deniker's data, all Russian races of the preceding 
correlation table being excluded, and I obtained the same negative result. As the 
total number of races in both tables is 176, this result is most probably correct. 
I had very few data for women— only 28 cases — and they give me a very slight 
positive correlation. The material is not large enough to draw conclusions 
concerning the sexual difference. It follows from the masculine correlation that 
the head-form changes when stature is selected. 
2. The correlation between the cephalic index and the shape of face, i.e. the 
relation of bizygomatic breadth to the total height of the face, measured from the 
top of the forehead to the chin, is positive and very sensible. It proves that the 
head-form can be influenced by sexual selection, because the shape of face is a 
prominent feature of racial " habitus." But as we pass to the upper face index, 
i.e. to the relation upper face height to bizygomatic breadth, which is calculated 
* "Ueber die Vererbung der Kopfform von Seiten der Mutter," Correspondenzhlatt der Deutschen 
Anthrop. Ges. No. 12, 1903. 
