294 



Messrs. Sorby and Butler on the Structure of [Feb. 18, 



the specimen is heated to the temperature of boiling water. Those con- 

 taining a similar fluid to that included in sapphires do occasionally occur ; 

 and when they are minute, they are extremely interesting, since they show 

 the spontaneous movement of the bubbles to greater perfection than any 

 mineral that has come under our notice. This is perhaps to some extent 

 due to the nature of the liquid, which is more mobile than the saline 

 aqueous solutions contained in the cavities of the quartz of granite and 

 syenite. It is manifestly a molecular movement analogous to that seen in 

 all matter when very minute particles are suspended in a liquid, so as to 

 allow freedom of motion ; and the rapidity of the movement is certainly 

 dependent on the size of the particles. It is not seen to advantage if the 

 diameter of the bubbles is more than of an inch ; but when it is about 



5T5WO tbey move to and fro in the most surprising manner, with such rapi- 

 dity that the eye can scarcely follow them. 



The number of small crystals of other minerals included in rubies is often 

 very great. There must be at least four different kinds ; but it would be 

 difficult to determine what minerals they all are. Some are very well cha- 

 racterized octahedrons, variously modified ; and, as shown by fig. 5, their 

 planes are very generally arranged parallel to planes of the ruby, and to 

 the small plate-like crystals already mentioned in describing sapphire. 

 These octahedrons have no influence on polarized light, and in general form 

 and character correspond so closely with spinel that it seems very probable 

 that they are that mineral. For some time we thought they were angular 

 fluid-cavities filled with liquid ; but when cut across in the sections they 

 are clearly seen to be solid, though less hard than ruby. Many of the 

 other included crystals are of such very rounded forms that, if it were not 

 for their action on polarized light, they might easily be mistaken for cavi- 

 ties filled with some fluid. Most of these rounded crystals are colourless ; but 

 some are of more or less dark orange-red colour, and are certainly not the 

 same mineral as the colourless or the octahedral crystals ; and in all proba- 

 bility the thin and' flat are a fourth kind. Occasionally alternating plates 

 of ruby with their axes in different positions gave rise to a beautiful series of 

 coloured stripes when examined with polarized light. 



Spinel. 



The ruby spinels from Ceylon sometimes contain fluid-cavities which 

 differ in a striking manner from those of any other mineral that has come 

 under our notice. One of these is shown in fig. 7. They are to a great extent 

 filled with a yellow substance, indicated by the shading, which seems to be 

 either a solid or a very viscous liquid. It incloses transparent, sometimes 

 well-defined cubic crystals, which have no action on polarized light ; trans- 

 parent, prismatic, or plate-like crystals, which strongly depolarize it ; and 

 black opake crystals, either in larger pieces or mere grains. The rest of 

 the cavity is in each case about one-third full of a colourless liquid, which 

 seems to contract on the application of heat, because it passes entirely into 



