OALCAEEOUS SKELETON OF TEE ANTHOZOA. 517 



indeed I have observed the small spicules from the body-walls 

 of the polyps of Spongod es to be enclosed in cells. All that 

 has been stated of the spicules of Spongodesis true, mutatis 

 mutandis, for the spicules of Siphonogorgia, and Kolliker's 

 figures of the spicules of Eunicea leave no doubt that the same 

 is true of this genus also. 



It is of importance to my present argument that the spicules 

 of the Alcyon aria show a definite and complex crystalline 

 structure, the details of which are, indeed, moulded upon and 

 dominated by an equally complex organic matrix, but are not 

 the expression of any particular arrangement of the cells from 

 which the spicules are formed. It cannot for a moment be sug- 

 gested that the radial cords and their warty emergences, or the 

 longitudinal bundles of crystalline fibres, are formations due to a 

 particular arrangement of calcified cells. The spiculoblast forms 

 an organic matrix, and secretes carbonate of lime. The latter 

 seems to be crystallised out, pari passu with the growth of 

 the former, and the crystalline growth conforms to the organic 

 growth, but that is all. 



I have elsewhere (4) given some details of the structure of 

 the skeleton of Heliopora coerulea. As this Alcyonarian 

 differs from all its allies in having not a spicular but a so-called 

 lamellar skeleton, it seemed to me that a renewed investiga- 

 tion of its structure might throw some light on the questions 

 raised by Mrs. Gordon's paper. Nor have I been disappointed. 

 The skeleton of Heliopora proves on closer examination to 

 be remarkably instructive, and to throw considerable light on 

 vexed questions regarding the Madreporarian skeleton. In 

 examining Heliopora I had recourse to flat and macerated 

 preparations of the growing tips of the corallum. These were 

 always stained with picro-carmine. I also made sections of 

 the growing tips after very careful decalcification with 1 per 

 cent, acetic acid, staining my preparations with picro-carmine 

 and picro-nigrosin or with Heidenhain's iron haematoxylin 

 followed by acid fuchsin. I cut some sections of hard and 

 soft parts together, the fragile and delicate character of the 

 corallum at tlie growing tips enabling me to do this without 



