2/4 C. M. CHILD. 



length. However much the small pieces may elongate and de- 

 crease in transverse diameter the tentacles remain "too large." 



Various other series of my experiments afford similar results, 

 but these two cases are sufficient as illustrations. 



Stated in general terms the conclusions drawn from the experi- 

 ments are as follows : as the size of the piece decreases the rela- 

 tive amount of regeneration increases - -provided of course that 

 the level at which regeneration occurs is the same in all cases. 

 It follows from this that the form, i. c\, the proportions, of 

 regenerated specimens may differ widely from the typical form 

 and that either reduction of the regenerated parts or extensive 

 new growth in the longitudinal direction must occur. Reduction 

 of the tentacles occurs, as has been noted (Child, '04^) but it is 

 due to change in external rather than internal conditions. If the 

 pieces are fed they will of course in time attain approximately 

 the typical proportions. But the important point is that the 

 regenerative processes in absence of food are not proportional. 



I am inclined to believe that these facts indicate that the stimu- 

 lus to regeneration is as great in small as in large pieces ; that 

 the difference in absolute amount of regeneration is due to the 

 earlier relative exhaustion of the smaller pieces, i. c., the absence 

 of energy available in the presence of a given stimulus ; and 

 that finally if this material could be supplied in such manner as 

 to be available only for the regeneration of a particular set of 

 structures, c. g., the tentacles, they would continue to grow until 

 equilibrium resulted between formative stimuli and destructive 

 factors, no matter what the size of the piece, provided it was 

 above the minimum. In other words, it is probable that the 

 decrease in the absolute amount of regeneration in Cerianthus 

 with decrease in size of the piece is due merely to decrease in 

 amount of available energy and not to any inherent capacity for 

 proportional regeneration. Furthermore, it is probable that vari- 

 ous other cases of so-called proportional regeneration are in 

 reality similar in nature to this case. 



REGULATIVE CHANGES OF PROPORTION IN THE OLD PARTS. 



The next subject which requires consideration is the change 

 of form of pieces as a whole. A cylindrical piece cut from the 

 body is of course relatively shorter and less slender than the 



