Di'. Odling on Ortho- and Meta-silicates. 369 



volatile anhydride, an ortbo-, nieta-, or intermediate phosphate 

 is produced, according to the facility with which the volatile an- 

 hydride can be driven off from its base. Thus I find that when 

 sulphate of magnesium is heated with phosphoric anhydride, a 

 tribasic or orthophosphate of magnesium is produced, M^ PO*, 

 sulphuric anhydride being somewhat readily expelled from mag- 

 nesia ; but when sulphate of potassium is fused with phosphoric 

 anhydride, we have only a monobasic or metaphosphate of po- 

 tassium produced, KPO^, in consequence of sulphuric anhydride 

 being very powerfully retained by potassa. The phosphoric acid 

 obtained from the magnesian salt gave a yellow precipitate with 

 ammonio-nitrate of silver, and did not coagulate albumen. The 

 phosphoric acid obtained from the potassic salt gave a white 

 precipitate with nitrate of silver, and did coagulate albumen. 



The case of silicic acid appears to be precisely similar to that 

 of phosphoric acid. In default of a correct knowledge of the 

 composition of siliciuretted hydrogen, we rely upon our acquaint- 

 ance with chloride of silicon. The vapour-density of this com- 

 pound requires the existence of four atoms of chlorine within its 

 two-volume molecule, and its formula is consequently Si CI*, 

 with which expression, moreover, its reactions best accord. Cor- 

 responding to this chloro -derivative of a tetrahydride of silicon, 

 we ought to have a class of tetrabasic, quadroxy- orthosilicates, 

 having the general formula M'* SiO*, and constituting the fourth 

 term of the following series : — 



Perchlorates . M CIO*. 



Sulphates . . M^ S 0*. 



Phosphates . . M^ P 0*(-M2 = M P 0^). 



Silicates. . . M'*SiO''(-j\PO = M2Si03). 

 Precisely as the tribasic are the most highly basic of the phos- 

 phates, so are the tetrabasic the most highly basic of the silicates ; 

 and as the tribasic orthophosphatcs, by the loss of an atom of 

 base, M^ 0, become monobasic metaphosphatcs, so do the 

 tetrabasic orthosilicates, by the loss of an atom of base, ]\I^ 0, 

 become biba.sic metasilicates ; and as there are intermediate 

 or pyrophosphates, with seven atoms of oxygen, so are there 

 intermediate | silicates with seven atoms of oxygen; and as 

 phosphoric anhydride, when fused with the salt of a volatile an- 

 hydride, may form an ortho-, meta-, or intermediate phosphate, 

 according to the facility with which the volatile anhydride is 

 liberated, so does silicic anhydride, when fused with the salt 

 of a volatile anhydride, yield an ortho-, meta-, or intermediate 

 Hilicate, according to the facility with which the volatile anhy- 

 dride is liberated, as in Colonel P. Yorke's experiments ; for 

 carbonate of lithium evolves carbonic anhydride more readily 

 than docs carbonate of potassium, while carbonate of sodium is 

 in this respect intermediate between the two. Hence with silica 

 Phil. Mng. S. 4. Vol. 18. No. 121. Nov. 1859. 2 U 



