NOTES ON REGENERATION. l6l 



piece is also removed, it will be found that the middle piece with- 

 out regenerating a new head will still regenerate a new tail. This 

 shows conclusively that Lillie's supposition m regard to posterior 

 regeneration is erroneous. The remainder of his argument, 

 which rests on this assumption, also falls, I believe, in the light 

 of this fact. 



The great mortality that Lillie observed in the posterior pieces 

 is due largely, at least in my experiments in which the same 

 thing was observed, to the temperature being too high, or possibly 

 to exposure to light. If the pieces are kept cooler (by sur- 

 rounding the dishes by the cool, running salt water of the labora- 

 tory) the mortality is much reduced, and instead of dying after 

 six days, as in Lillie's experiment, I have kept short posterior 

 pieces for several weeks. It is only by keeping such pieces for 

 a long time that one can fairly test their powers of regeneration. 



Schultz states that he cut Dendroccslum in two either between 

 the pharynx and the reproductive region or else in front of the 

 pharynx. In the former case he found that the posterior pieces 

 regenerated an anterior end very slowly, and he found it more 

 profitable in studying the regeneration of the head to use those 

 posterior pieces that had been cut off in front of the pharynx. 

 He found that the regeneration of the anterior end often failed to 

 take place, and he attributes this to fusion of the sides of the 

 cut surfaces, as I had found to occur not infrequently in Bipaliinn. 

 Whether this is the whole of the question remains to be seen. 

 In a marine polyclad, Leptoplana, Schultz found that posterior 

 pieces, no matter at what level they have been removed, fail to 

 regenerate an anterior end, even when only a small piece of the 

 head is cut off. Yet regeneration from a posterior cut surface 

 takes place at all levels. Schultz attributes the lack of regener- 

 ation at the anterior end either to the closing over of the " grow- 

 ing point" by the coming together of the old tissue from the 

 sides, or to the muscles from the sides uniting and thus prevent- 

 ing further growth. Both factors he thinks may enter into the 

 result. This point could be tested, I think, by making the cuts 

 so that there is left a pointed anterior end, when regeneration 

 should occur, if Schultz's view is correct. From an experiment 

 of this sort that I have carried out on Dendroccelum I think it 



