522 GILBERT 0. BOURNE. 



clearly how the component crystals are arranged along the edge 

 of such a spine; fig. 23, which is a surface view, shows the 

 arrangement of the crystalline components on a larger scale. 

 Both drawings might be taken for representations of an 

 ordinary inorganic crystalline deposit. In fig. 22 the crystal- 

 line elements are seen to diverge, more or less regularly, from 

 a centre of growth, and this is the usual condition; but in 

 fig. 23 the components are arranged pell-mell, the long axes of 

 the crystals crossing one another at every angle : only at the 

 right-hand edge is there an indication of a more regular diver- 

 gent growth from a centre. 



Although the structure of the corallum of Heliopora is 

 much coarser than that of any madrepore which I have 

 examined, it is difficult to isolate the crystals, so as to 

 identify the crystal system to which they belong. On crush- 

 ing, the corallum breaks up into fragments, some of the 

 most regular of which are represented in fig. 24, a, b, and 

 c. The smallest fragments, such as b and c, behave under 

 polarised light like single crystals, being regularly extin- 

 guished in four positions 90° apart when rotated between 

 the crossed Nicols. Treatment with dilute acids breaks 

 up the crystals into needle-like fibres resembling those of 

 Spongodes, and every such fibre behaves as a single crystal 

 under the polariscope. 



There is no evidence whatever that the crystalline elements 

 are formed within cells. In my previous paper I gave a 

 drawing of a section through the corallum at the tip of a 

 growing frond of Heliopora. The arrangement of the crystal- 

 line fibres was only indicated by faint diverging lines. By 

 preparing thinner sections, and examining them with higher 

 powers and polarised light, I was able to make out the dis- 

 position of the crystalline elements sufficiently clearly. The 

 coarser structure of Heliopora makes it a good object for 

 study. Fig. 25 is a representation of a portion of a transverse 

 section magnified 420 times. The first thing to be noticed is 

 the absence of dark lines or '' centres of calcification,^^ such as 

 occur in Madreporaria, and have been so well figured and 



