Form Regulation in Harenactis attenuata 373 
tacles should not be formed about the circumference of the oral 
region of the ring, provided the physiological characteristics of 
this region remain after closure the same as they were before. 
The only possible inference from the facts as we know them at 
present seems then to be that the union of oral with aboral end has 
altered, at least quantitatively, the physiological characteristics 
of the oral end. Concerning the aboral end the case is not so 
clear in the earlier stages of restitution, though, as will appear 
below, the aboral end is apparently altered more than the oral 
end, since it gives rise in many cases to tentacles. 
In short, a decrease, though apparently not an elimination of the 
original polarity in the piece constitutes the first step in determin- 
ing what follows, 1. e., the establishment of a new polarity, for we 
cannot doubt that each of the tentacle groups represents the estab- 
lishment in some degree of a new oral-aboral axis. 
We have now to consider how the regions where this new polar- 
ity may be established are localized. In the first place conditions 
are not identical about the whole circumference. When the piece 
is cut out from the parent body the mesenterial organs are extruded 
as an irregular mass of tissue. The very crude and indefinite 
operation of removing this mass by a transverse cut as indicated 
in Fig. 2 must accomplish the same general result as an operation 
along the broken lines in Fig. 31 would accomplish, provided 
such an operation could be performed, when the animal was in the 
distended condition. The two lines a and b, Fig. 31, serve to 
indicate roughly the results of the removal of less or more of the 
mesenterial organs, as this result would appear if the piece could 
regain a fully distended condition. It is evident from the figure, 
and examination of the pieces also shows very clearly, that the 
mesenteries are most completely removed in the terminal regions 
of the piece. Frequently the mesenteries are practically wholly 
removed at some regions of the circumference near the cut end, 
more frequently the oral end, since extrusion is usually greater 
there. In such cases nothing but the body-wall remains. But 
the results of the operation are never uniform about the whole 
circumference, at least not in the cases under my observation, and 
it is not likely that they can be in any given case. At some points 
