964 THE AMERICAN NATURALIST. [Vor. XXXIV, 
hooks in the same colony ; or, in other words, whether the 
variability in the number of hooks found on statoblasts of the 
same colony will be less than the variability of the statoblasts 
from all the colonies taken together. To answer this question, 
I selected the eleven colonies showing the greatest number of 
statoblasts, and determined for each colony the e and the c. 
Then I averaged the e's and the c’s for the eleven colonies, 
and compared these averages with the e and the c of the 
635 statoblasts coming from known colonies considered 
together. If the average e or c of statoblasts from the same 
colony is no less than the e or c of the population at large, there 
is no heredity. If, on the other hand, the es and c's of the 
colonies average zero, so that the differences between the vari- 
ability inside the colony and in the whole population is the 
greatest possible, then there is the strongest possible heredity. 
In general, the percentage which the difference between the 
averaged e or c is of the e or c of the whole population, sub- 
tracted from unity, will serve as an index of heredity. I find: 
€ E 
Average for 11 colonies, 1.197 8.772 
All statoblasts from colonies, 1.326 9-597 
: ; 1.197 
Index of relative heredity based one, 1 — 1226 = I — 0.9027 = .0973. 
8.772 
“ & és [1 i e re p = 140 .0860. 
9.597 : 
The hereditary tendencies inside the colony are thus seen 
not to be very strong when compared with the tendency to 
similarity between all the statoblasts of the complex of colo- 
nies. I interpret this to mean that there is a close relation- 
ship between all the colonies in the complex. 
The question next arose, what is the ontogenetic explanation 
of this variation? 
To get some light on this matter, I studied the method of 
development of the hooks. As already stated, the hooks arise 
at or near the outer margin of the ring-shaped float ; there is a 
little variation in the line of origin. Even after the float has 
been formed, the statoblast is clad externally with cells. Those 
at the outer edge of the statoblast are narrow and elongated 
