ORDERACTINOIDEA. 53 



The secretion of a calcareous lamella or plate takes place between 

 two visceral lamelke, but whether from the surfaces of these fleshy 

 lamellfe, or from a prolongation inward of the membrane forming 

 the walls of the visceral cavity, has not been ascertained. The secre- 

 tion does not always commence at the central plane of the plate, but 

 often either side of it; for they are sometimes hollow within. In some 

 species the spines of the surface are tubular, as in the Echinopora 

 reflexa. We have observed, with regard to the visceral lamelke, ex- 

 hibited in figure 15, ^ 25, that a large compartment alternates with 

 three narrower ; and also tliat the large calcareous lamellfe in the Den- 

 drophyllise and many other species, alternate with three smaller : it 

 appears therefore that the larger rays of a cell are formed between a 

 pair of large visceral lamella?, and the smaller in the narrower spaces 

 which are intermediate. 



The calcareous secretions forming these corals are often deposited at 

 successive intervals or as layers in the zoophyte. In a Madrepora the 

 surface between the cells becomes covered with minute points by the 

 continued secretions, and then a layer forms, connected with the pre- 

 ceding by these points or columns. The interior usually becomes, 

 afterwards, nearly solid by additional secretions. This variety of 

 structure may be observed also in the Dendrophylli;e ; and even 

 the compact species, in which there are no traces of cellules, will 

 often show evidence of having been deposited in layers. I have seen 

 it brought out with singular distinctness in a specimen half fossilized, 

 the layers easily peeling off from one another. In many corals, how- 

 ever, we fail to detect this deposition in layers. This is the case in 

 the Astrpea tribe. The Pocillopora?, and some allied corals, have 

 transverse plates crossing the cells internally, which are intermitted 

 secretions from the lower part of the polyp; but no appearance of layers 

 has been detected in the spaces between the cells. The Favosites 

 and many Cyathophyllidse are examples of similar interrupted secre- 

 tions across the cells. 



47. Our explanations thus far, apply more especially to the Acti- 

 naria. The Alcyonaria secrete no rays within the cells, nor even 

 striate the inner surface of the same, while rays or striating lamelte 

 belong to the cells of all the Actinaria. Moreover, in the Alcyonia 

 group, the secretions, instead of forming layers, constitute dissemi- 

 nated granules or spicule, some of which are represented in figure 22, 

 and there is every gradation, from the purely fleshy species to those 

 which are quite firm, from the secreted calcareous grains. These 



14 



