CALCAREOUS SKELETON OF THE ANTHOZOA. 511 



examined with the highest powers. The distinction between 

 the coenocytial layer and the cell layer is, perhaps, of no great 

 importance, but it is of more importance to note that the 

 arrangement of the crystals composing the spicule bears little 

 relation to the structure and arrangement of the cells from 

 which the spicule is formed. 



The actual crystalline structure is but slightly indicated in 

 the fresh spicule, but it becomes evident after treatment witli 

 dilute acids. A spicule which has thus been partially decalci- 

 fied or " etched '^ is seen to be composed of a number of minute 

 calcareous fibres which radiate from the centre towards the 

 periphery. Their arrangement, however, is not perfectly radial. 

 They cross and overlap one another to some extent, and many 

 of the fibres crop out on the surface of the sclerite, and do not 

 reach its edge. Hence there is some amount of interference, 

 and the dark cross shown in fig. 7 is not perfectly symmetrical, 

 nor are all parts of its limbs equally opaque. Some of the 

 crystalline fibres may be isolated at the edges of the etched 

 sclerite. Each is a minute acicular crystal about 0*03 mm. 

 long, and behaves under polarised light as a true uniaxial 

 crystal. 



I did not succeed in isolating the organic constituents of the 

 spicules of Primuoa or Plumarella. 



Thirdly, there is the spindle-shaped spicule in its many 

 varieties, usually covered with spines, warts, and projections of 

 different kinds. My observations on this form of spicule were 

 almost entirely confined to Spongodes and Siphongorgia 

 mirabilis, whose spicules are of comparatively simple form, 

 and large enough to admit of my making transverse and longi- 

 tudinal sections. 



The spicules of Spongodes (sp. incert.) are shown in fig. 9. 

 They are long fusiform structures, covered with warty pro- 

 jections and of very variable length. Some of the larger 

 spicules, which serve as supports for the individual zooid heads, 

 are as much as 3'5 mm. long; the smaller, which lie in the. 

 ■walls of the zooids, are about 0*2 mm, long. They are 

 variously coloured, crimson, orange, or pale yellow, according 



