SECT. 2] 



AND WEIGHT 



457 



the diameter of the eye, the length of the head, and the length of the 

 caudal fin. Fig. 60, taken from Jenkinson, shows how, although the 

 curves do not run absolutely parallel, they certainly rise and fall 

 together. On the other hand, the correlation coefficients between 

 several pairs of organs show that in many cases — total length and 

 breadth of caudal fin, total length and length of anterior dorsal fin, 

 total length and length of head, head length and eye diameter — 

 there is a significant diminution in value during the time that there 



70 



a 

 .^ 60 



iB 



50 



40 



C 



« 30 



c 



y 20 



o 



10 



.3 1'3 2'3 3*3 4*3 5-3 6*3 7'3 8*3 9*3 10-3 



Weeks after hatching 



Fig. 60. 



is a decrease in growth-rate. Thus, when the growth-rate is highest, 

 the variations between individuals are greatest, but the correlation 

 coefficients between various organs or parts in the same individual 

 are also greatest. It is easy to see why this should be so, but Boas 

 has given a mathematical proof of the relation between these co- 

 efficients and the growth-rate. Expressed differently, it could be 

 said that the faster the growth-rate the more proportional the growth 

 but the greater the variation as between individuals. 



Turning now to the early part of the embryonic period, the first 

 complete investigation was that of Fischel, who studied the individual 

 variations between duck embryos at the primitive streak stage. Von Baer 



