1910-11.] Study of the Tasmanian, Australian, and Papuan. 25 
The coefficients of variation for length, breadth, and height for the 
three races show (see Tables IX. and X.) that as regards length and 
breadth the minimum figure occurs in the Tasmanian and the maximum 
in the Papuan, whilst as regards height the maximum figure occurs 
in the Australian and the minimum in the Papuan — results which in 
the main show that the greatest uniformity of type occurs in the 
Tasmanian, with the Papuan at the other end of the scale ; or, in other 
words, that the Tasmanian is the purest of the three races compared, 
the Papuan the least pure, and the Australian intermediate. 
These results are more strikingly brought out by reference to the 
coefficients of correlation, which we regard as of primary importance 
in the present investigations, because they are functions of shape only. 
The relative value of these coefficients as criteria for comparative purposes 
has next to be considered. 
There is not, unfortunately, much information available as to what 
importance is to be attached to the comparison of breadth-height and 
height-length, but of the importance of the comparison of length and 
breadth every craniologist is fully convinced, for it is to the varying 
relationships of these dimensions that are due the transitions from dolicho- 
cephaly to brachycephaly. 
This general observation is confirmed by the results of Pearson and 
Lee’s data (27), which is, according to Hatai (28), the only work as 
yet available for comparisons of the coefficients of correlation of breadth- 
height and height-length. In general, their work shows that the correla- 
tion for length and breadth and for height and length is about the 
same, but that both are much higher than for breadth and height. 
For the albino rat Hatai (28) obtains precisely reverse results ; but, 
as he shows, “the difference is probably due to differences in relative 
development of the several bones of the cranium, depending on the 
skull-form characteristic for the two species.” 
Tschepourkowsky (29) has also shown that, with increase of the body 
length, head length and head height also increase, but that head breadth 
and facial breadth decrease. From race to race the breadth diminishes the 
greater the length formation of the head, but the height increases. His 
figures finally show that in the different races of mankind the head length 
is less variable than the breadth, as is also the height, and hence form 
differences are chiefly due to increased breadth, as may be clearly seen by 
comparing the ranges of the mean lengths, breadths, and heights in our 
Tables IX. and X. 
It is thus sufficiently clear both on biometric and on morphological 
