No. i .] RECENT EXPERIMENTS ON REGENERA TION. 3 1 



might be secured, but from the general tenor of the experi- 

 ments I should doubt whether anything less than a full quarter 

 of the body would reproduce a new individual. This, however, 

 is merely inference based on the general facts observed. Spe- 

 cific experiments to be undertaken later will possibly show 

 very different results. 



In the third series the experiments consisted in vertical 

 sections through the median portion of the body, dividing it as 

 nearly as possible into halves, as indicated in Fig. 3, and with 

 the result that each half became 

 an independent and perfect me- 

 dusa. In this case the restora- 

 tion was somewhat peculiar. It 

 would seem to be a sort of re- 

 covery of form and function rather 

 than regeneration in the usual 

 sense of that term. After the 

 first brief shock of the operation, 

 which in many cases was scarcely FlG 



noticeable, was overcome, there 



was an evident effort of the half-medusa to assume the normal 

 form by a contraction of the body so as to bring the cut edges of 

 the bell together, the approximations of the edges taking place 

 from above downward, or peripherally, and with the subsequent 

 union or healing of the approximated margins. Complete 

 restoration, resulting in the assumption of the original form, 

 occurred in from three to five days. The new medusae were 

 in most respects quite similar in form and action to the origi- 

 nal, though of course of only about half the size. The time at 

 my disposal was insufficient to observe whether there was sub- 

 sequent growth of the specimens. In the recovery of the 

 specimens I was -not able, moreover, to observe any disposition 

 to regenerate the additional radial canals necessary to complete 

 the symmetry of the original. This, however, does not seem 

 to be an important matter, since there does not seem to be a 

 special necessity for a definite number. And in this connec- 

 tion it may be worth while to note a very considerable varia- 

 tion in normal forms as to the number of radial canals. 



