466 Capt. W. A. Ross on Pyrology, or Fire Analysis. [June 20, 



Microscopically viewed, however, these crystals are very beautiful, especially 

 in polarized light ; and it seems at any rate certain that silica and the sili- 

 cates thus treated invariably crystallize in most elegant and beautiful 

 arborescent appearances, the form taken by other salts being usually that 

 of a disk, often of a leaf. 



Cobalt, Copper, and Metallic Occides. 



73. The behaviour of these in B has not yet been fully examined ; but 

 such as have, promise results quite as interesting and important as those 

 which may be derived from that of the earths. For instance, when an ore 

 containing the oxides of cobalt and nickel, previously manipulated as in para- 

 graph 90, is boracically treated as described in paragraph 61, the cobalt im- 

 mediately congests into globules, which after 0. P. appear blue-black, after 

 H . P. red-purple through a lens ; the nickel oxide, on the contrary, remains in 

 amorphous fragments, which are bright green after O. P., and have a metallic 

 lustre after H. P. Cupric oxide forms in O. P. balls of an indigo-blue colour, 

 not easily distinguishable at first from those of cobalt; but in H. P. the cu- 

 preous globules immediately give out streaks of the red suboxide, which can- 

 not be mistaken, though, of course, were further proof necessary, the addition 

 of 5 per cent, of soda would form a pink glass from the first, and a blue one 

 from the second. Iron oxide remains in amorphous fragments of a black- 

 brown colour with a rusty halo or tinge round them, and is thus easily 

 distinguished. Oxide of Uranium forms a stringy black mass with a yel- 

 lowish opacity round, which the addition of soda dissolves to a bright pea- 

 green bead. Molybdic acid affords many curious and beautiful changes, 

 for a description of which there is not space. 



Silver, Lead, and the Volatilizable Oxides. 



74. None of these form either balls or fragments in boric acid, but 

 spread over the whole bead as a milky precipitate, that of silver having a 

 slight pinkish tinge. Nothing, therefore, can be easier than their separation 

 thus from those metallic oxides which form balls or fragments, as the former 

 of these can not only be collected or aggregated into one sphericle, but ex- 

 tracted from the cold bead with thegreatest ease, as described in paragraph 63. 

 It is obvious that this process would probably form an important method of 

 extracting silver from its ores with very little loss; for the boric acid pro- 

 tects oxides contained in it from the direct effect of an O. P., which dissi- 

 pates pure silver unprotected to the extent of 10 per cent, in a very short time. 



75. Altogether, as a detective reagent, boric acid seems scarcely in- 

 ferior even to phosphoric acid, while as a separating one it is quite un- 

 surpassed. 



Many hypotheses of the formation of these spherieles or balls have sug- 

 gested themselves to the writer, but none with a sufficient weight of evi- 

 dence in their favour to be stated here. They may be due {a) to capillary 

 phenomena, (b) to the retention of a certain amount of carbonic acid (for of 



