4 HARG1TT. 



in the laboratory, while in most of the others it was taken directly 

 from the open harbor. Since in these two sets the medusae were 

 placed in smaller dishes, the consequent overcrowding may have 

 been the cause of their death, through lack of aeration. 



In order to ascertain whether the regeneration of new tentacles 

 would take place more rapidly if a portion of the margin was 

 left, than when it was entirely removed, three quarters of the 

 margin was excised leaving one quarter of the original number of 

 tentacles still present. The part from which the tentacles had 

 been removed gradually contracted till it occupied only about 

 one quarter or one half of the new margin. Thus two or three 

 canals were crowded into less than half of the margin and two 

 of them sometimes fused as shown in Fig. 3. New marginal 

 canals were fully formed in about a week, and the circulation of 

 the fluid in them could be observed. In ten or twelve days small 

 bulbs were found on the margins of some, and from these bulbs 



FIG. 3. FIG. 4. 



tentacles formed, while at about the same time new otocysts devel- 

 oped. As suggested under the second set these bulbs are prob- 

 ably aggregations of new tissue, and are perhaps comparable to the 

 tentacle " anlagen " in regenerating hydroids. Fig. 4 shows the 

 portion of the margin from which the tentacles were excised, o 

 represents otocysts and / the new tentacles. It will be seen that 

 the pads or bulbs b found at the bases of the old tentacles are lack- 

 ing in the new tentacles at this stage, and form only after the new 

 tentacles have attained considerable growth. This fact has been 

 previously observed and noted by Hargitt l 1901. The tentacles 

 1 BIOLOGICAL BULLETIN, Vol. II., p. 244. 



