166 
Proceedings of the Royal Society of Edinburgh. [Sess. 
wise of the Australian and Tasmanian to a somewhat severe proof. Whatever 
may be the result, it must be remembered that it is the first time that such 
an attempt has been made on the lines of strictly severe scientific analysis ; 
and further, that in submitting the question to such proof, we are now 
enabled to deal with the largest numbers of Australian and Tasmanian 
crania which have ever yet been employed ; and lastly, that the application 
of such an analysis to what previously has been mere theory is due to the 
introduction of some ingenious craniological methods by Dr Th. Mollison, 
formerly of Zurich, and now of Dresden. 
In 1908 Mollison published in the Zeitschrift fur Morphologie und 
Anthropologie his “ Beitrag zur Kraniologie und Osteologie der Maori ” (9). 
In this paper he introduced for the first time what he then termed the 
“ Abweichungsindex.” The object of what in English we should term the 
“ variation index ” is to discover if two skulls belong to one and the same 
race or not, and the procedure adopted is as follows : — 
Multiply the distance of the individual from the average value of the 
standard type group by 100, and divide the product by the sum of the 
extremest distance of the group on the minimum or maximum side of the 
variation breadth. Thus a standard group of skulls (see fig. 3) has an 
average greatest breadth value of 135, and a cephalic index average value 
of 76. Another skull to be compared with this group has a greatest 
breadth of 126 and a cephalic index of 64. The question to be answered 
by the variation index is, Does this skull belong to the same race as the 
standard group or not ? The distance of the doubtful skull from the 
average value of the standard group is for the greatest breadth 9, and for 
the cephalic index 12. Multiply these figures, both of which are on the 
minimum range of variation side of the standard group, by 100, and divide 
by the greatest range of variation of the standard group on the maximum 
or minimum side as the case may be : in this case the minimum side. The 
correct figures will therefore be found in the example quoted by subtract- 
ing for the greatest breadth 129 from 135, and for the cephalic index 70 
from 7 6. In each instance the difference is 6. If this calculation be worked 
out in the manner indicated, it will be found that the variation index for 
the imaginary object to be compared with the standard group is, for the 
greatest breadth 150, and for the cephalic index 200. 
To compare the variation indices of a large number of characteristics, 
draw a straight line which shall be supposed to pass through the average 
figures of each characteristic. Parallel to this draw two lines at arbitrary 
distances and supposed to represent the minimum and maximum range of 
variation of each observation from the average for the same; in each 
