M. Marignac on the Silico-tungstates. S77 



nese. It dissolves in hydrochloric acid with disengagement of 

 chlorine, yielding a solution of a clear saffron-yellow colour, which 

 diminishes on dilution, and disappears completely on boiling. 

 Nitric and sulphuric acids dissolve the peroxide with a reddish- 

 yellow colour, forming corresponding compounds of the oxide. 



Popp finds that the peroxides of nickel and cobalt are formed 

 in the same manner as that of cerium. 



Marignac* has found that when a solution of acid tungstate 

 of potash or soda is boiled with gelatinous silicic acid, the liquid 

 assumes an alkaline reaction, and is now found to contain a new 

 acid in which 1 equiv. of silica is united with 12 equivs. of tung- 

 stic acid, SiO 2 12 WO 3 , which he calls silico-tungstic acid. It 

 is a powerful acid, which is easily separated from its salts, forms 

 two splendidly crystallized hydrates, and salts which for the most 

 part are soluble and crystallize well. 



Under the same circumstances as above, acid tungstate of am- 

 monia gives rise to another acid in which an equivalent of silica 

 is united with ten of tungstic acid, SiO 2 , 10WO 3 , and which 

 Marignac names silico-decatungstic acid. It is more difficult to 

 obtain pure ; its hydrate does not crystallize ; and from the very 

 soluble character of its salts, it is difficult to obtain them in 

 crystals. 



The acid itself is not very stable ; when heated, a small quan- 

 tity of silica separates out, and a new acid is obtained containing 

 silica and tungstic acid in the same proportions as in silico- 

 tungstic acid, but which in other respects is totally different, and 

 which has accordingly been named tungsto- silicic acid. It forms 

 a soluble and deliquescent hydrate, which can, however, be ob- 

 tained in well-formed crystals. It forms a series of salts isomeric 

 with the silico-tungstates, but differing from them in crystalline 

 form, and in the amount of water of .crystallization, and also in 

 being somewhat more soluble. 



All these acids are quadribasic, if the salt with 4 equivs. base 

 is to be taken as neutral, and are formed when the acids act 

 upon the carbonates of a base. The most frequent salts are 

 those with 2 or 4 equivs. base j the first crystallize most easily. 

 They have a great tendency to form double salts ; hence it is 

 that ammonia does not precipitate a solution of silico-tungstate 

 of alumina, and that alumina, magnesia, carbonate of lime, &c. 

 are dissolved by silico-tungstate of ammonia. As the salts of the 

 acids are easily soluble, and contain a large quantity of tungstic 

 acid, their solutions have a remarkably high specific gravity. 

 The solution of silico-tungstate of soda has the specific gravity 

 3*05, so that glass, quartz, and most rocks float upon this, which 

 is moreover a very mobile liquid. 



* Comptes Rendus, vol. lviii. p. 809. 



Phil. Mag. S. 4. Vol. 29. No. 197. May 1865. 2 C 



