34 
PEOFESSOR K. PEARSON ON THE INFLUP:NCE OF NATURAL 
than the Aino. The variabilities S., and Sj of humerus and radius for a population 
selected from the French by femur and tibia are obtained from (Ixiii.) and (lix.). 
We have : 
' S 
= -urn + T81 [ ^ ) + -268 ) +2 X -221 
/ -S ■ 
■Ns. ' -3 
.) h _3 
cr., 
S. 
Sp/o-fr = -354 + -108 -- ) + -205 ^ 
These give for the particular case : 
2 in 
O-T (To 
Pi3 X -169. 
^3= 1-39, 
-r 
ITl. 
Turning to the correlation of humerus and radius, we have by (lx.) : 
cr., cr, ^ / S 
hi = S y -'20 + -140 ( - 
-3 —I L ^ 
1 / 
+ -267 ( 
S.. 
(To 
+ -389 -pi, ^ - 
giving for the particular case : 
1*34 — *799. 
It will thus be seen that if we selected from the French a group with the same 
variabilities and correlation of femur and tibia as the Aino, the variabilities and 
correlation of the humerus and radius of this grou|D would not be very different from 
those of the Aino. On the other hand, the correlations between upper and lower 
members would be very significantly different. 
(Tenerally we have by (Ixi.) for selection from the French : 
r„ = yV426^ +-518p^>. 
A 4 
v 
—4 
_ N J -q.^Q A 
9 + -515 p 
a 
r..3= .518-4 4. -423 
~ ^ ' To ■ ' T, 
T., 
l-o, =: -515 + .329 p3_ 
— 4 
To 
These yield for our })articular case : 
ti3 “ *819, 14 ^ 
l.o.j — 845 , 
•694, 
•768. 
These are all smaller than the corresponding French values, the selection has 
reduced the correlation, but the Aino po]3ulation has in all the cases but >’03 a greater 
