Form Regulation in Harenactis attenuata 277 
for the localization of definite tentacle groups about the circum- 
ference of the ring except as I have indicated above. 
The other incidental factor is the presence of a layer of tenacious 
slime over the surface of the body. Harenactis, like Cerianthus, 
secretes this slime in greater or less amount according to condi- 
tions. Inthe rings it is particularly annoying because it frequently 
encloses all parts of the ring. In my experiments I removed it 
as often as possible but it was usually secreted anew. Cylindrical 
pieces readily creep out of it but the rings are incapable of this. 
Very frequently this slime prevents the distension with water of 
tentacles which have formed, and if this condition is allowed to 
continue these tentacles may undergo complete atrophy. The 
atrophy of all but one tentacle of the group c in Figs. 24 and 25 
and of the single tentacle at d in Fig. 24 (Cf. Fig. 25) was due to 
this slime. The single tentacle which persisted in the group c 
protruded through an opening. It seems probable therefore that 
this slime may in some cases prevent the appearance of tentacle 
groups by mere mechanical pressure. In a number of cases 
where the rings were left for a week or more without removal of 
the slime they became completely enclosed in it and no tentacle 
groups appeared. But while this factor like distension may pre- 
vent the appearance of tentacles or groups or may bring about 
atrophy of those already developed it is only an incidental factor, 
for the tentacles appeared in groups in cases where the rings were 
kept practically free of slime. 
3 The Establishment of the New Polarity and Symmetry 
Our conception of the nature of the processes involved in the 
establishment of a new polarity will of course depend on our con- 
ception of the nature of polarity. If we regard it with Driesch as 
a directive property or characteristic inherent in each ultimate 
particle, element, molecule, or whatever we choose to call it, of 
protoplasm, we must conceive a change in polarity as involving 
a coordinated change in direction of the axes of these particles. 
If we adopt Morgan’s hypothesis that it is the expression of an 
