580 bulletin: museum of comparative zoology. 



form the connectives is probably independent of any stimulus due to 

 contact with the digestive tube. (3) In normal regeneration a lump 

 or cone of new tissue develops at the regenerating end. This enlarges 

 and grows outward, ultimately forming normal segments. The 

 experiments show that this outward growth is not conditional upon 

 pressure exerted by the alimentary canal. (4) It is conceivable 

 that the stomodeum in a regenerating worm is formed by a backward 

 pull exerted, by some mechanism inside the body, upon the epidermis 

 covering the regenerated tissue. But the end of the digestive tube 

 is certainly not a necessary part of any such mechanism. (5) Also 

 the brain and connectives are eliminated as possible mechanical 

 agencies in the formation of the stomodeum, for in one animal (num- 

 ber 359) the latter appeared before the brain and commissures had 

 developed. 



4. Summary. 



1. Following the excision of the anterior region of the nerve cord 

 and digestive tube from a beheaded earthworm, these parts may 

 regenerate, thereby restoring the normal structure of the head. 



2. The regeneration of the alimentary canal and nerve cord is 

 independent of mechanical stimuli or support provided by either 

 cicatrix or regenerating body wall at the anterior end of the worm. 



3. What may have been an abortive attempt to regenerate a head 

 on the dorsal side of the body was observed in worms from which the 

 anterior regions of both nerve cord and digestive tube had been 

 removed. 



4. The development of a stomodeum is not contingent upon the 

 presence of brain and connectives. 



5. The following structures can regenerate independently of any 

 mechanical stimulation or support furnished by the digestive tube: 



a. Stomodeiun. 



b. Connectives and brain. 



c. A region of segmented body wall containing longitudinal and 



circular muscles. 

 I wish to express my appreciation to Prof. H. W. Rand for propos- 

 ing the problem, for giving me many suggestions in the course of the 

 investigation, and for assistance in preparing the manuscript. I also 

 wish to thank Prof. E. L. Mark for kindly criticisms in revising the 

 paper. 



