82 C. M. Child 
characteristic of either pole of the axis may occur, but heteromor- 
phic development of oral or anterior structures is probably of more 
frequent occurrence than aboral or posterior heteromorphosis; 
but this is probably due, however, to the fact that the oral or an- 
terior region ise region of greater physiological activity than the 
other, 1. e., negative results in aboral or posterior pieces are often 
due rather to lack of energy than to anything else. 
In general heteromorphosis of oral or anterior structures occurs 
in regions near the oral or anterior end, heteromorphosis of aboral 
or posterior structures nearer the opposite end. In certain cases, 
however, as in Tubularia, axial heteromorphosis of oral structures 
may be a frequent or characteristic result at most or all levels of 
the body. Here and in all cases of this kind the external condi- 
tions after isolation are a factor in determining the result, conse- 
‘quently such cases do not afford data which can be used directly 
for the analysis of polarity (Child ’07). They tell us nothing about 
an original polarity except that, if it ever existed, it is no longer 
present. In fact, as I pointed out in my discussion of polarity 
in Tubularia (Child ’07e) we must distinguish two kinds of axial 
heteromorphoses, primary and secondary. Primary heteromor- 
phoses are the necessary result of the axial organization of certain 
regions which existed in the original organism. ‘They occur in 
all cases where the piece is so short axially that axial differentiation 
or specification is almost or wholly absent. In the absence of the 
original organization the parts cannot be correlated with each other 
as are the parts of the whole. The two terminal regions, being 
physiologically alike or nearly so, react similarly to the existing 
conditions, with little or no relation to each other, consequently 
new specifications are established with reference to these two 
regions instead of one of them, and we say that the polarity of 
one end has been reversed. ‘The heteromorphic discs and ten- 
tacles in cesophageal pieces of Harenactis are undoubtedly primary 
heteromorphoses, i. e., there are no marked axial or polar physio- 
logical differences in this region in the original animal, and when 
a piece is isolated new polarities arise between each end and the 
middle, since conditions must necessarily be different in these 
two regions. Similarly the posterior heads in short pieces of 
