Form Regulation in Harenactis attenuata 379 
general terms the dominant region of later stages. Doubtless 
‘many complicating factors exist in different cases, but they do 
not alter the fact that polarity reduced to its lowest terms may con- 
sist merely in a quantitative difference in reaction or in certain 
reactions along some axis. 
If these suggestions are correct, it is evident that a polarity of 
some kind will arise wherever such differences in reaction or in 
reactive capacity are established along an axis. If such a region 
remains in correlation with the original whole so that the nature of 
its reactions is determined by its correlations, the result will be a 
polarity or symmetry of parts or organs. If, on the other hand, the 
correlations with other parts of the original whole are reduced or 
eliminated and if the cells involved in the new relations are toti- 
potent, 1. e., possess the same capacities or “‘potences”’ as the ovum, 
then a new system or individual must of necessity arise if life con- 
tinues, and its principal axis must coincide in direction with the 
differences in reaction rapidity which constituted the starting 
point. 
This very brief consideration of the problem is perhaps tnsufh- 
cient to show clearly the basis for my conclusions, but it must serve 
for the present. I hope to present a more complete statement of 
my position and the grounds for it at another time. 
Returning now to the problem of the establishment of the new 
polarities about the circumference of the rings, certain points sug- 
gest themselves onthe basis of whathasbeensaid. Inthe first place, 
if regions of more rapid growth are established at certain points in 
consequence of greater injury to the old system, as suggested above, 
a gradation in physiological activity between the areas of most 
rapid or most intense reaction and the less active regions about 
them must exist. I believe that this gradation represents the 
first step in the establishment of the new polarity. If the original 
correlations between the various parts still persisted these regions 
might undergo further development as parts of a whole, but we 
have seen above that union of the oral with the aboral end has 
altered these correlations, at least quantitatively, i. e., the original 
polarity has been decreased or almost eliminated. Consequently 
the different regions grow independently of each other, at least 
