vii EXPERIMENTS ON BE ROE 361 



pole, and after the movements had again become regular and 

 vigorous in both the oral and aboral portions of the rows, but 

 before their connection in the two portions was completely 

 restored, the animal was completely divided into two parts 

 by deepening the cut already made, then the movement in 

 the two parts immediately after the complete separation went 

 on exactly as it had before the separation. In such a case 

 the larger tube-shaped oral portion of the animal swam away 

 exactly as if it were an entire uninjured animal, just as a 

 branch of an oleander tree, which has been made to send 

 forth roots while still attached to the stem by being sur- 

 rounded at a particular spot with earth contained in a pot, 

 goes on growing independently when it is separated from the 

 mother-plant by a cut below the roots. 



Thus, while the movements of the swimming -plates is 

 stopped for a time by dividing a Beroti into two parts or 

 into several, and usually also by merely making a superficial 

 cut in the animal, yet when the separation is preceded by the 

 preliminary operation above described, it may be completed 

 without producing any effect on the divided portions. This 

 proves that the two portions acted quite independently before 

 the complete separation, and that the reappearance of the 

 continuity of the movement of the plates must be the con- 

 sequence of a subsequent concrescence, or must be caused by 

 the establishment of new nervous paths which take the place 

 of the old. 



Facts are also known which prove the vicarious action of 

 nerve-fibres in Vertebrata, for example, in the results which 

 follow from alternate section of the two halves of the spinal 

 cord. 



The facts above brought forward not only show how 

 rapidly during individual life parts of the organism can be- 

 come adapted to special functions ; they enable us also to 



