ZOOLOGY OF THE BERMUDAS. Ill 



more or less foliaceous fronds towards the periphery; a sphinc- 

 ter of the diffuse type occurs upon the inner surface of the 

 disc between the inner tentacles and the outer tentacles or 

 fronds ; and the stomatodseum is provided with two deep goiiid- 

 ial grooves, which are prolonged some distance below the 

 inner extremity of the stomatodseum. 



The family Phyllactidse was placed by Andres in the sub- 

 order (family) Stichodactylinse, the fronds being considered 

 homologous with tentacles. I have here ventured to remove 

 the family to the sub-order Actininse, and it will be necessary 

 to furnish my reasons for such a change. The tentacles must 

 necessarily be considered outgrowths of the disc, since struct- 

 urally they resemble it closely while differing greatly from 

 the column. Are the fronds also disc structures? 



The question turns upon what we shall consider to be the 

 limit between the disc and the column. The majority of 

 authors have taken a more or less distinct fold of the body 

 wall, the margin, frequently furnished with conspicuous 

 acrorhagi, to be the boundary, and certainly in many cases 

 there seems to be a marked difference on either side of this 

 fold. Thus, the column may, as in Bunodes and Phymactis, be 

 turberculated as far as the margin, but beyond this the tuber- 

 cles cease, and there is apparently a decided difference between 

 the region below and that above the limiting fold. 



In the Sagartidse and Paractida3 there is imbedded in the 

 column wall below the margin a sphincter muscle. In other 

 forms, however, such as the Bunodidse, which possess a circum- 

 scribed endodermal sphincter, that structure lies internal to the 

 margin. If we assume with the Hertwigs that the sphincter 

 is a columnar structure, its situation in the Bunodidse would 

 indicate that the margin is not the boundary between the disc 

 and column. 



Neither the margin nor the sphincter, however, can be con- 

 sidered the morphological boundary of the disc, since both seem 

 to vary somewhat in position. The true criterion is to be found 

 iu the difference of histological structure presented by the disc 



