198 Prof. S. J. Hickson. On Ceratopora, the [June 15, 



The tubes do not communicate with one another below the surface, and 

 there are no tabulae. 



The walls of the tubes are brown at the surface, but this brownness 

 gradually fades away, as the walls are traced downwards, into a pure white 

 marble colour. This difference in colour is due, I believe, to a difference 

 in chemical constitution as the walls grow older and thicker. 



An examination of the vertical fracture further shows, when it is highly 

 magnified, a number of long and very slender tuberculate spicules, partly 

 imbedded in the walls and partly projecting on the surface and into the 

 cavities of the tubes. All these spicules are arranged vertically, that is to 

 say, parallel with the long axis of the tubes (figs. 3 and 4), and they project 

 upwards into the cavity of the tubes as the latter widen out towards the 

 surface. When a group of two or three tubes are broken off and placed in 

 dilute nitric acid, the free projecting parts of the spicules rapidly dissolve ; the 

 lower parts of the walls of the tubes also dissolve in the course of a few hours, 

 but the upper, free, and brown parts of the tubes remain for several days as 

 a soft flexible substance, in which the basal parts of the spicules may be seen 

 until they are dissolved. 



My interpretation of this experiment is that the walls of the tubes, as they 

 were formed at the surface, were composed of a horny organic substance, in 

 which a few long spicules of calcium carbonate were imbedded ; in the lower 

 and older parts the horny substance became impregnated with calcium 

 carbonate, and finally, at the base, nearly the whole of the horny organic 

 substance became replaced by the inorganic salt. 



The method of formation of the crystalline calcium carbonate is not very 

 easy to understand, and, the specimen being unique and of small dimensions, 

 I have not felt justified in making more than a few sections and other pre- 

 parations. From these, however, I feel satisfied that the construction of this 

 coral lum is on very similar lines to that of the corallum of Heliopora as 

 described by Bourne.* There are vertical trabecular from which the crystal- 

 line rods diverge in three directions, meeting in sutural junctions with similar 

 diverging systems. These vertical trabecular can be traced for some distance 

 down into the solid subjacent parts of the cap. There are no dark lines or 

 centres of calcification such as occur in the Madreporaria. On crushing a 

 very small fragment it breaks up into short irregular angular rods, very similar 

 to the fragments of Heliopora drawn by Bourne in his fig. 24. From- the 

 consideration of these observations it seems quite probable that, as in 

 Heliopora, the corallum of Ceratopora is formed by " crystallisation of 

 carbonate of lime in an organic matrix." 



* 'Quart. Journ. Micr. Sci.,' 1899, vol. 41, p. 499. 



