294 Experimental Zoology 



stock, or the likeness or unlikeness of the united levels, may 

 account for the results. 



A third possibility also exists in grafts made in this way. The 

 graft itself may swing around into line with the trunk of the stock, 

 and become the head of the new hydra (Fig. 16). The head of 

 the stock shifts down to the base and there pinches off to pro- 

 duce a new hydra, leaving its original posterior end in the posses- 

 sion of the new head. 



If a very short piece of the anterior end is grafted near the 

 head of another hydra (Fig. 17), the two heads fuse into one. 

 At first there are too many tentacles, but some of these are ab- 

 sorbed or even two tentacles fuse into a single one a process 

 not uncommon in hydra, producing, while in process of comple- 

 tion, the forked tentacles not infrequently found. 



If the cut end of a hydra is grafted into the side of another 

 hydra, and then, after union, the graft be cut off close to the 

 stock (Fig. 1 8), the small ring (whose outer end closes) will be 

 slowly absorbed into the stock. The result may be expressed 

 in terms of my hypothesis as follows : The resistance in the cells 

 of the small piece is insufficient to allow the piece to pinch off, 

 and it is too different to permit it at first to share the common 

 trunk. Not being able to free itself, and unable to maintain 

 itself under unfavorable conditions of tension, it is absorbed, 

 or changed over into a part of the body wall, the process 

 being very slow as a rule. 



It has been found 1 that different species of hydra behave 

 somewhat differently. In Hydra fusca, lateral grafts tend to 

 move forward until the head end is of the same length as that 

 of the stock when fusion begins and unites the two parts. If, 

 however, the lateral graft is inserted less than one fifth the dis- 

 tance from the lower end, it moves downward and constricts 

 off at the base. In Hydra viridis, lateral grafts tend to move 

 downward wherever united to the stock, and separate at the 

 foot, unless inserted very near the head end of the stock, when 

 the two heads fuse. 



1 Hefferan (1902). 



