af an nd 
FRETS, THE INDEX CEPHALICUS. 523 
capacity and index and for German brachycephalic skulls the 
same author finds a positive correlation. MACDONELL (1904, p. 374) 
calculates for his English dolichocephalic skulls a coefficient of 
correlation, that shows that there is no sensible correlation be- 
tween capacity and index (r — —.02). PEARSON (1904 p. 374) 
publishes a table that shows that the relationship between capacity 
and index is slight and irregular. He states in accordance with 
LEE that the correlation of capacity and index is negative for 
dolichocephalic races and positive for brachycephalic races. 
In a dolichocephalic race (Lee’s Aino skulls) the larger heads 
are consequently more dolichocephalic (have a lower index). The 
greater capacity is reached by augmenting the length and we know 
(p. 512., tab. 14 and 15) that a greater length is connected with a 
relatively smaller breadth and consequently with a lower index. 
In a brachycephalic race (LEE’s Bavarian skulls) the larger heads 
are more brachycephalic. Here the greater capacity is reached by 
augmenting the breadth and we have seen, that a greater breadth 
is connected with a relatively smaller length and consequently 
with a higher index. - ; 
Grouping material of a dolichocephalic, resp: a brachycephalic 
races we shall always find of course that grouping according to 
engthclasses the index decreases and grouping according to 
breadthclasses the index increases. 
Finally tab. 20 (p.052) makes it possible to compare my results 
with those of others, that have reported on cranial material in a 
standardised form. 
As has been said at the beginning of this communication (p. 481) 
my investigation respecting the heredity of the shape of head led 
me to a further analysis of the index cephalicus. 
As a result of the correlation-investigation (p. 511 and p.521) 
it appears that, if the breadth of head grows larger, also the 
length of head will generally increase, but in a less degree. Now 
it is the question what is the correlated height of head in this 
case. I performed this investigation for the material of TOCHER. 
As for the material of men, we find for the height-iength index 
that, with an increasing length the index will fall, and with an 
increasing height it will rise (tab. 21 and 22). 
We also find for the height-breadth index that, with an increasing 
