FORM-REGULATION IN CERIANTHUS. 



195 



tractile and changeable in form as Cerianthus the correspondence 

 between the levels of cuts on different specimens is always only 

 approximate. In this case the levels at which the oblique cuts 

 were made differ more or less in the three specimens as well as 

 the levels of the transverse cuts in the controls. Any control 

 is therefore only approximate and the pieces used for this purpose 

 here are as satisfactory in this respect as those employed in other 

 experiments on Ceriautlius. 



In the control pieces A and B the rapidity of regeneration is 

 that characteristic of the level of the body at which the oral ends 

 of the pieces lie. Comparison of this rapidity with that of the 

 extreme levels of the oblique cut surfaces will show whether the 

 obliquity of the cut has any effect on the rapidity of regeneration. 



FIG. 3. 



FIG. 4. 



After the operation the collapsed oblique pieces acquired a form 

 like Fig. 3. During the next day or two the cut edges rolled 

 inward still more closely and the pointed oral portions became 

 more or less bent over toward the cut surfaces. All became 

 more or less flattened as they lay upon the flat glass bottom of 

 the aquaria. 



September 15. 3 days after section. Experimental pieces : 

 One piece with a few minute tentacle-buds at the uppermost por- 

 tion of the oblique cut surface ; others without tentacles. 



Controls. All without tentacles. 



September //. 5 days after section. Experimental pieces : 

 All the pieces were completely or nearly closed orally by new 

 tissue : the aboral ends were not completely closed by new 

 tissue but were closely approximated and two of the pieces were 

 slightly distended. Each of these pieces possessed a small group 

 of minute marginal tentacle-buds at the extreme oral part of the 



