536 GILBERT 0. BOURN K. 



there was a desmocyte, such as is shown in fig. 42, conspicuous 

 among the calicoblasts for its dark colour, but of tracts of 

 desmocytes no trace whatever. 



In the section under consideration the soft tissues have 

 shrunk away from the corallum. At tlie edge of the latter the 

 striated scales (Mrs. Gordon's calcified calicoblasts) are clearly 

 seen, and adherent to tlie raesoglcea, and lying on the corallum 

 where the latter is in contact with the soft tissues, are calico- 

 blasts — irregularly-shaped rounded cells, and decidedly smaller 

 than the calcareous scales. That the calicoblasts appear in 

 the section as separate cells, and not as a continuous layer of 

 protoplasm, is probably due to shrinkage. In other parts of 

 the section, where the soft tissues have not shrunk away from 

 the corallum, the calicoblast layer appears continuous. The 

 rubbed-down sections are too thick for satisfactory examination 

 with an immersion lens, and one cannot speak with certainty, 

 but in decalcified sections from the same species the calicoblast 

 layer always has the characters represented in fig. 42. 



The other point noticeable in the section is the fact that the 

 surface scales on the same level as the polished upper surface 

 of the section appear to be emergences of tracts of fibres 

 which diverge from a centre of calcification which is not 

 included in the drawing. Mrs. Gordon's drawings of sections 

 of Madrepora are represented on a much smaller scale than 

 mine (loc. cit., figs. 60 — 63), but in her fig. 63 there is a 

 distinct indication of the same fact. Now if her theory were 

 true, viz. that each scale represented a calcified ectoderm cell, 

 it would follow as a necessary corollary that the calcified cells 

 were piled upon one another in perfectly regular series, cell 

 fitting accurately on the top of cell, and in such a manner as 

 to produce the regular diverging columns shown in my figure. 

 This is extremely improbable, as anybody well acquainted with 

 the arrangement and behaviour of cells will allow. And a 

 comparison with the structure of the Spongodes spicule shows 

 that such an assumption is unnecessary for the explanation of 

 the structure. In the Spongodes spicule the ground substance 

 is made up of regularly disposed concentric lamellae, and each 



