46 Mr. W. Thomson on the 



glass are the potassium sodium and calcium silicates, which 

 are not acted upon by the strong sulphuric acid above 

 mentioned, and which are developed from the surface of the 

 glass by the slow solvent action of the fluorides, just as the 

 metallic tin crystals are developed from the surface of tin 

 by the solvent action of dilute hydrochloric acid upon its 

 surface. 



The objection to this theory is that glass does not 

 polarize light; but it cannot be deduced from that that 

 glass is not crystalline, because Pasteur proved that although 

 paratartaric acid does not polarize light it is still crystalline, 

 and is composed of crystals of the irregular system, but 

 that the crystals or molecules are so arranged that the 

 polarizing influence of one is neutralized by the reverse 

 action of another always found in juxtaposition with it. 

 Is it not possible, then, that glass crystals may be simi- 

 larly arranged to each other so that the polarizing influence 

 of one crystal may be neutralized by the reverse polarizing 

 influence of the other? And this seems to be borne out 

 by the fact that whilst small sodium and potassium silico 

 fluoride crystals shew distinct polarization when viewed 

 simply by two Nicol's prisms, the crystals on the glass do 

 not shew polarization by that means, and it was only by the 

 employment of circularly polarized light, produced by 

 passing the light through a quartz plate, that a distinction 

 could be observed between the crystals in question and 

 ordinary glass. I believe that these crystals are then 

 silicates of potassium, sodium and calcium, etc., and that 

 they are not produced by the combination of the solvent 

 with some of the constituents of the glass. Ammonium 

 fluoride, when heated on the surface of glass, developes a 

 beautiful fern-like structure on it resembling hoar-frost on 

 a window pane. 



