SoLLAS — On sponge Spicules. 389 



The angle of extinction for the spicule will be about 20° (17° 

 in the projection, figure 7), and for the strise a like amount. 



It is obvious that if the rhombohedron occupy this position, one 

 will, on rolling the spicule under a cover-glass, alter the observed 

 index of refraction and the angle of extinction. 



The reason that any approach to constancy in these elements 

 is discoverable in the spicules, as they chance to lie on the micro- 

 scope slide, is due to their bow-shaped form and lateral compression, 

 so that they rest most readily on one side ; and the fact that, under 

 these circumstances, a close approximation to constancy occurs, 

 shows that the crystalline structure of the spicule is related in a 

 more or less constant manner to its form. 



On rolling the acerate spicules of Grantia and those of Leu- 

 conia glomerosa (?) under a cover-glass, the expected changes 

 were observed ; the refractive index varying from about its mini- 

 mum to about its maximum value for the extraordinary rays, 

 and the angle of extinction from 0° to 44°. In some positions 

 the colossal spicules gave a refractive index similar to that of 

 the sagittal triradiates, i. e. they vanished in balsam ; the strise 

 should then extinguish at 0°, and make an angle of 75° with 

 the edge of the spicule, which should consequently extinguish 

 at 15°. These conditions were found on measurement to be rigor- 

 ously fulfilled. 



We may fairly conclude, then, that the acerate spicules of those 

 calcispongise examined are built up of excessively elongated pri- 

 mitive rhombohedrons of calcite, which lie with the optic axis in-: 

 clined at about 75° to the morphological axis of the spicule. 



That this result holds good not only in spicules which have 

 been treated with acid, but in unaltered spicules taken from 

 spirit specimens (and we may infer in living examples), is shown 

 by the similar behaviour of both when examined with polarized 

 light. 



The position of the rhombohedron relative to the surface of the 

 spicules is very similar to that which may be observed in rhombo- 

 hedra of calcite filling up a cavity within a limestone rock, or 

 inside the chamber of an ammonite ; and we may suppose that the 

 deposition of calcite within the spicule sheath occurred according 

 to just the same laws as those which are followed in the purely 

 mineral world; the first formed layer of minute rhombohedra 



