22(i University of California Publications in Zoology. [Vol. c 



the migration of endodermic cells one by one into the Erustule 

 occurs on polyps of all sizes, that method of establishing the axis 

 has been abandoned in adult regeneration of the tentacles, an 

 intrusion of a considerable group of cells in several rows lead- 

 ing more rapidly to the establishment of the structural char- 

 acteristics peculiar to the tentacles of polyps the size of the 

 reg( aerate. 



I purposely avoid using the term "future significance," in 

 this connection, since it possesses distinctly unfortunate con- 

 notations. The interpretation of the method (if development 

 exhibited by a tentacle is to he sought, not in the result of the 

 process, bul in the physiological conditions present in the 

 regenerating piece and governing the process of regeneration. 

 It is unfortunate to speak of a development proceeding 

 toward a given end. Rather is the development proceeding in 

 accordance with conditions existing in the developing region. 

 with no po*ssible reference to any end. I formulate this point 

 of view here not for a novelty which it does not possess, hut 

 because it is so well supported by the facts obtained from Cory- 

 morpha. As will be shown more fully in another paper, 8 regen- 

 erating pieces of Corymorpha shape themselves to a greater or 

 less degree according to their length and diameter. In short 

 pieces there may be very striking decrease in the diameter of the 

 column and in the complexity of its structure. With this 

 obvious simplification of structure, this retrogression toward the 

 degree of differentiation characteristic of the larva, there is a 

 corresponding decrease in the size of the tentacular fundaments 

 and the tentacles developing from them. Short pieces do not 

 always undergo a large degree of simplification. In such eases. 

 the size of the tentacles is proportional to the diameter of the 

 regenerating piece (which represents a certain degree of differ- 

 entiation), not to its mass or volume. 



If it is possible to make the facts obtained from Corymorpha 

 the basis of a generalization, it would appear that the phenomena 

 of regeneration are controlled primarily neither by considera- 

 tions of future significance nor geometrical relations, but by the 



s See also Aspects of Regeneration in Corymorpha, Proc. 7th Int. Zool. 

 Congress, 1910. 



