378 C. M. Child 
axial gradation of materials then changes in polarity consist in 
changes in this gradation. But neither of these hypotheses gives 
us anything more than a formal statement of the fact of change of 
polarity. Moreover, as I attempted to show in the preceding - 
paper (Child ’ogb) neither of these hypotheses accounts for all the 
phenomena of polarity, though I believe that Morgan’s hypothesis 
is much more nearly correct, so far as it goes, than Driesch’s, for 
there is no satisfactory evidence in favor of a directive organization, 
while there is abundant evidence to show that specification or 
differentiation along the axis exists. 
Correlation as well as constitution is an essential factor in 
organic polarity; in fact under certain conditions differences in 
correlation may determine differences in constitution, 1. e., the 
constitution of a given region may be altered in one direction or 
another according to the character of its correlations with other 
parts. 
It can, I think, be shown that the phenomena of organic polarity 
are possible in a system or individual in which the only localized 
differences of any kind are differences in the rapidity or intensity 
or both of reaction along some axis, but these phenomena are }os- 
sible only when correlations exist. In such a system one region 
will react more rapidly or more intensely to a given stimulus than 
others and the correlative effect of this reaction, must necessarily 
be the alteration of the reactions in other parts. Moreover, the 
processes in the region of greatest reaction rapidity or intensity 
must produce a greater effect upon other regions than the processes 
in these regions produce upon it. In other words it becomes the 
dominant region of the system physiologically, and in the following 
structural and functional complication which must occur in such 
a system it is in advance of other parts and may determine more 
or less completely their fate, according to the nature and degree of 
correlation. I am inclined to believe that the ovum is essentially 
such a system, that in its simplest form the only “polarity” present 
may be a difference in the rapidity (or intensity) of reaction along 
some axis. It is evident that in very many cases, if not in all, the 
region of greatest reaction rapidity at the beginning of develop- 
ment, 1. e., the “animal pole”’ becomes the head region or in more 
