Form Regulation in Harenactis attenuata 391 
While we are far from a complete analysis of these reactions and 
the conditions upon which they depend, there is every reason to 
believe that they are as characteristic under certain conditions as 
the “normal” reaction is under others. 
But whatever the particular factors involved, the rings show how 
readily polarity and symmetry may be altered or obliterated and 
new polarities and symmetries arise. Moreover, they afford an 
excellent illustration of the readiness with which an organic sys- 
tem breaks up into smaller systems when the old correlations be- 
tween parts are obliterated. The union of parts into a whole 
seems to me to be essentially a matter of correlation: if this is the 
case, the whole must cease to exist, at least physiologically, when 
the correlations are eliminated or altered beyond a certain limit. 
But whether, under these conditions, new wholes shall or shall not 
be formed, will depend first on the capacity of potence of the ma- 
terial and second on the local conditions in various regions. If, 
for example, local conditions determine in totipotent material, 
the establishment of a new polarity in any region or regions anew 
whole or wholes may arise. Probably, as was suggested above 
in discussion of the establishment of the new polarities, polarity 
in its simplest form may arise in consequence of very simple or 
very slight local differences in living material. Whether such 
polarity becomes the polarity of a whole or of a part will depend 
likewise on the character of the material involved and the environ- 
mental conditions (internal correlations and sometimes external 
conditions) which are present. 
In the rings it is evident that new polarity and symmetry are 
established, but in consequence of gradual collapse in all cases, 
growth does not continue sufficiently far to show whether the new 
systems would finally reach a condition of equilibrium similar to 
that of the old, or differing from it more or less widely. If a ten- 
tacle group with a mouth should appear on one of these rings it 
would probably be possible to keep the whole complex alive 
indefinitely. 
Those parts of the tentacle groups which arise from tissue 
belonging to the original aboral end of the piece (Cases I to V, 
Figs. 8 to 20) are of course axial heteromorphoses; moreover, 
