112 THE BERMUDA ISLANDS. 



and colamii ectoderm. This layer in tlie disc possesses ecto- 

 dermal muscle-cells and a nerve-layer, which structures are 

 absent in the column. The tentacles resembling the disc in 

 structure are to be considered outgrowths of that region, and 

 passing outward from these one finds that the characteristic 

 structures of the disc gradually fade out and are lost. It is 

 impossible to say just where the change is completed, but the 

 region in which it occurs must be considered the boundary be- 

 tween the disc and column. In Bunodes tseniatus and Aviactinia 

 steUoides I find that the sphincter muscles lie beneath the outer 

 border of this indifferent region, and are consequently to be re- 

 garded as columnar structures. 



In the Phyllactidas the sphincter muscle lies between the 

 tentacles and the fronds, and although the ectoderm in the re- 

 gion in which it occurred, and in the area between the fronds 

 or their representatives and the margin was completely mac- 

 erated away in the forms studied, j'^et reasoning from the rela- 

 tions of the sphincter in other forms we must conclude that the 

 region between the margin and the base of the tentacles is 

 columnar, and that the fronds and- outer digitiform tentacles 

 are column structures, perhaps comparable to acrorhagi, and 

 cannot be considered homologous with tentacles. Accordingly', 

 only one tentacle belongs to each intra-mesenterial space, and 

 the Phyllactidas must be referred to the sub-order Actiniuse. 



Andres, in the introduction to his Monograph, notes the fact 

 that the margin does not always mark the boundary between 

 the disc and the column. He proposes the term " collar " to 

 denote the portion of the column internal to the margin. 

 Gosse's term " fosse " is not applicable in all cases, as for in- 

 stance in Condylactis, where the region does not form a depres- 

 sion, but is horizontal. 



Oulactia fasciculata. n. sp. (PI. 10, fig. 5.) 



By this name I denote three specimens in various degrees of 

 contraction, the largest of which measured about 1 cm. in 

 height and 1-2 cm. in breadth. The color, as ascertained from 

 alcoholic specimens, is in the lower part of the column a gray- 



