Eaymond Pearl 
83 
12. The correlation of braiu-weight with stature and with body-weight is 
positive and in each case the regression is linear. 
13. The weight of the cerebrum follows the same laws of vai'iation and cor- 
relation as does the weight of the entire brain, but exhibits a somewhat greater 
relative variability. 
14. The results as to variation and correlation in brain-weight are closely 
accordant for all the races studied. 
15. The indirect selection of bodily characters can modify the mean weight of 
the brain to a considerable extent. The amount of this effect has been measured 
in certain cases. 
16. There is no evidence that brain- weight is sensibly correlated with intel- 
lectual ability. The limits of this correlation have been shown to be not closer 
than 0 and + '6. 
17. So far as can be determined from present material dolichocephaly has 
associated with it no advantage in mean brain-weight over brachycephaly and 
vice versa. 
18. There is great need foi- further large and homogeneous collections of 
brain-weight statistics. When these are available for a considerable number of 
races it will be possible to pass from intra-racial to inter-racial problems. 
APPENDIX OF MEASUREMENTS. 
Correlation Tables, 
The fundamental correlation tables on which the results of this paper are based 
have been placed together here at the end for convenience in reference. The 
tables are numbered consecutively in Arabic numerals, and each bears its own 
explanation. A word may be said as to the plan of arrangement. The racial 
order is 
Swedes Tables 1 to 12 inclusive. 
Hessians „ 13 „ 24 „ 
Bohemians „ 25 ,, 34 „ 
Bavarians ,, 35 „ 44 „ 
English „ 45 and 46 „ 
Under each race the order of arrangement is in each case for the tables 
involving brain-weight as one of the variables to precede the subsidiary tables 
involving only the related characters. The tables of males precede those for 
females, and the " total " series precede the " young." 
11—2 
