SOLITARY CORALS. 233 



Vox H eider counted eighteen tentacles in one-half of a specimen of D. ramea. 

 My specimens of D. gracilis were too much contracted to enable me to count with 

 certainty, but I found, both in the bud and in the adult, indications of a cycle of 

 exotentacles alternating with the primary and secondary endotentacles. The exo- 

 tentacles appear to be simply introverted, and do not enter into close relations 

 with the mesenteries. The endotentacles, as in Heteropsammia, are introverted and 

 doubled over the inner edges of the endosepta. As far as I could determine, there 

 ai*e twelve exotentacles in the adult, alternating with the primary and secondary 

 endotentacles. Thus there would be thirty-six tentacles of both kinds, and my 

 observations agree with von Heider's. 



The stomodajum is relatively longer than in Heteropsammia, measuring fully 

 1 millim. in length in the contracted spirit specimen, and in the expanded polyp it 

 is probably much longer. I could find no trace of sulcus or sulculus. The stomodseal 

 ectoderm extends some little way down the free edges of the primary and secondary 

 mesenteries, as in Heterocyathus and Heteropsammia, but not so far as in these two 

 genera, and eventually it gives place to a normal mesenterial filament. 



As regards the histology, I have not very much to add to what has been published 

 by vox Heider. In D. gracilis the ectoderm of the body-wall is peculiar, and 

 unlike anything that I have seen in any other coral. It is difficult to obtain a clear 

 idea of its structure in sections, but it appears, as shown in fig. 26, that the cells are 

 large and vacuolated, and form a thickened cell-wall, which, from its staining 

 properties, seems to be of the nature of an intercellular substance. The walls of 

 adjacent cells cohering together give a semi-cartilaginous consistency to the whole 

 tissue, which no doubt forms an efficient protection against the numerous sponges 

 and other organisms that infest the majority of corals. Embedded in the ectoderm 

 cells are nematocysts of three kinds : ( 1 ) the common spiral-thread nematocyst 

 measuring 0"02 millim. in length, more or less ; these are very scantily distributed. 



(2) Medium-sized elongated nematocysts, about 0'028 millim. long and 0"005 millim. 

 broad (fig. 27b) with a loosely and irregularly coiled thread ; these are abundant. 



(3) Large elongate oval nematocysts about 0'035 millim. long and O'Ol millim. broad. 

 According to Gardiner (21) the medium-sized nematocysts are developing 

 stages of the spiral - 02-millim. nematocysts, and this may be the case. It is not 

 easy to speak with certainty on this point from a study of sections only, but there is 

 some reason to think that the medium-sized nematocyst, as I have figured it, is 

 really a third variety. In the first place they are very abundant in the ectoderm of 

 the body-wall, whereas the small spiral nematocysts are very scantily represented 

 there ; in the second place they occur in the stomodaeum in which no other 

 nematocysts are to be found. Von Heider found two kinds of nematocysts in the 

 ectoderm of D. ramea, the small spiral 0'02-millim. variety, and the large elongate 

 oval variety, measuring in this case - 05 millim. in length. He describes the latter 

 as tilled with a coarsely granular material, but without a thread, and a large 



2 H 



