1824.] M. Berzelius on Fluoric Acid. 451 



in which a portion of the silica is separated and replaced by 

 some other basis, which may be either an alkali, an earth, a 

 metallic oxide, or even water, it is incapable of entering into 

 combination with any other substance except a neutral rluate. 

 I have kept the gas in contact for many days with the pulverised 

 carbonates of potash and soda, without the slightest absorption 

 ensuing ; and a similar result was obtained with lime and with 

 the bicarbonates of potash, although I expected that the water 

 of crystallization of this salt might have facilitated the combina- 

 tion. On the contrary, the pulverised fluates of the alkalies, 

 earths, and metallic oxides, even when anhydrous, instantly 

 absorb the gas, and become saturated with it at the end of a few 

 hours. Consequently the fluoric acid and silica, when asso- 

 ciated in the proportions which constitute the gas, have no ten- 

 dency to combine with an additional quantity of base ; and the 

 class of salts styled fluosilicates, instead of being combinations 

 of a fluate with a silicate, are combinations of fluate of silica 

 with the fluates of different bases. 



QuantitativeCompositio?i of Fluate of Silica. — The silica preci- 

 pitated by water from this compound is so voluminous, and is 

 moreover so decidedly soluble in water, that it is impossible to 

 conduct an analysis by this process with any degree of precision. 

 The most advantageous method appeared to be to precipitate 

 the fluoric acid and silica by soda, in the state of the difficultly 

 soluble salt, and afterwards to separate by double decomposition 

 the silica which remains dissolved in the liquid ; but this pre- 

 supposes a knowledge of the double salt, which would obviously 

 render a separate analysis of the gas superfluous. After ascer- 

 taining this necessary preliminary, I proceeded to the analysis 

 in the following manner : A quantity of water was impregnated 

 with the gas, until it became converted into a thick coaoulum. 

 During the process, the liquid was incessantly agitated, and 

 care was taken to prevent it from ever coming in contact with 

 the conducting tube. The mixture was now thrown upon a 

 filter, and the silica was washed until the filtered liquid ceased 

 to redden litmus paper. Ignited, it weighed 1-263 gramme. 

 No trace of fluoric acid is expelled during the ignition. I con- 

 sider it necessary to mention this circumstance, because Gay- 

 Lussac and Thenard's experiments might lead to the supposi- 

 tion, that the silica separated by water from the gas is in combi- 

 nation with a smaller proportion of fluoric acid. The filtered 

 acid liquid was mixed with carbonate of soda as long as it effer- 

 vesced, and the double salt which precipitated was collected 

 upon a balanced filter, washed, and dried in a balanced platinum 

 crucible. It weighed 8-99 grammes, and is equivalent to 

 8*053 grammes of fluoric acid, and. to 2*994 grammes of silica, 

 provided we make the calculation from my former number for 

 the atomic weight of that substance, which, however, is a little 



2 <; 2 



