REGENERATION IN A TROPICAL EARTHWORM. 361 



AUTOTOMY. 



Several references have already been made to disturbing factors 

 which interfered with the success of some of the experiments. 

 One of factors, the most important, was a tendency for the worms 

 to break into fragments in early hours after the operation. For 

 want of a better term this process of fragmentation will be re- 

 ferred to as autotomy. Only very rarely was this fragmentation 

 observed to occur later than the first twenty four hours after the 

 operation, and then only very small portions usually consisting of 

 one or two segments were thrown off. 



In the first series of operations on P. excavatus many of the 

 operated worms autotomized one or more pieces from the posterior 

 portion. In another series of anterior operations thirty two out 

 of forty two animals autotomized portions of the tail ranging from 

 seven to sixty millimeters in length. No series of operations in 

 which tail portions of the worm were watched was free from 

 this tendency to fragment. In the majority of cases one or two 

 short pieces were autotomized from the posterior end of the major 

 operated portion. Such fragments were usually dead when dis- 

 covered but very often lacked the pungent odor so characteristic 

 of decaying earthworm. Occasionally the fragmentation was 

 much more striking and extensive. Worm B 9 from which eleven 

 anterior segments had been removed broke into pieces. Worm 

 D 19 from which X anterior segments had been removed broke 

 entirely into pieces six to ten millimeters in length. Several other 

 specimens from which eight or nine segments had been removed 

 also autotomized extensively. 



Autotomy is usually understood to be a throwing off by the 

 animal of a small portion which usually dies without producing 

 a new animal but in P. excavatus apparently any fragment from 

 any region may survive, or more than one of the fragments may 

 survive, with the survival determined by the presence or absence 

 of something in the worm and not by the position of the frag- 

 ment along the axis of the animal. The autotomy was observed 

 only in posterior portions. The length of the tail however was 

 of no significance. Posterior portions from one third to approxi- 

 mately nine-tenths of the length of the original worm autotomized 



