THE HYDRANTHS OF HVDROID POLYPS. 97 



which has been forced periodically from one end to the other of 

 the hydroid-colony, may, by distending the remains of the polyp, 

 delay absorption for a number of hours. If, however, the pres- 

 sure of the current is not great, the polyp grows gradually 

 smaller until only a small ball of material is left in the cup, and 

 this is then drawn down into the stalk. The whole process may 

 occur in six hours, or may be prolonged for two days or more. 

 A study of the prepared material shows that changes begin 

 first in the endoderm cells of the body of the polyp, into the 

 cavity of which are thrown fragments of degenerating endoderm 

 and gland cells. This continues for some time, and is accom- 

 panied by the contraction of the supporting lamella, as a result 

 of which the ectoderm changes from a flattened to a columnar 

 form. The cells of the hypostome round up rapidly at a com- 

 paratively late stage, and are set free into the digestive cavity ; 

 the lamella contracting as before. In the tentacles the endo- 

 derm is also in process of degeneration, and later, when a break 

 comes in the lamella at the base of the tentacle, the cells pass 

 into the body cavity. The ectoderm cells in this region are 

 thrown into folds which, seen from the surface, might easily 

 give the effect of being fused, as noted by Loeb ; but I have not 

 seen any signs of real fusion only many cases where the cells 

 of different tentacles are brought into close contact. During this 

 time the lamella of the tentacle breaks, and masses of nettle 

 cells and ectoderm pass through the break into the digestive 

 cavity. The broken ends of the lamella now draw together and 

 form a hollow shell, which is frequently much distended by the 

 pressure of the digestive current on the elastic lamella. Degen- 

 eration continues by the slow turning in of ectoderm and endo- 

 derm cells, until only a small fraction of the original polyp re- 

 mains, and this is then drawn through the opening at the base of 

 the cup. There are no signs, either external or internal, of any 

 drawing back of protoplasm to form a part of the stalk previous 

 to the final stage, but at this time the strands of protoplasm con- 

 necting the ccenosarc and perisarc at the end of the stalk are 

 broken, and in sections the masses of nettle-forming cells, \vhich 

 usually lie in the ectoderm just below the polyp, can be seen to 

 have moved farther down the stalk. 



