in the Formatibn of Saltpetre, 337 



lieat rendering the mass lighter, forces it to rise higher and 

 higher, and to attain the regions where the rarefaction of 

 the air £,ives to the electric flnid the Uberty of moving up 

 and down in the cohrnin; 



If M. SaussLire succeeds by this reasoning in accounting 

 for the loud detonations which we then hear, it does not 

 appear to me that it explains the causes oF the formation of 

 waltr, orof the other aqueous substances which are precipi- 

 tated. Mav we not be allowed to suppose that the positive 

 or vitreous flnid and the negative Or resinous fluid, being 

 brought into contact by means of the column of vapours, 

 feffeet the formation of the nitric acid; and that the water 

 held in solution by the azotic and oxygen gas, being set free 

 by their combination, occasions the phaenomena which we 

 observe in these circnnistances ? 



But if this formation of the nitric acid in the air were 

 doubtful, I could support it by a phajr.omenon generally 

 known, but which lias not vet received a satisfactory ex- 

 planation ; whereas, according to my theory, it is a fact 

 perfectly natural; and which, 1 must also add, has even 

 served as the basis of all my reasoning. 



We cannot conceive how the nitrate of potash comes to 

 be formed under the tiles of a roof, since this fact is in 

 contradiction with everv thing that we know on the su!)ject 

 of the nitrates; for, as [ have said, the nitric acid is only 

 formed, in our climate, in places where the air is stagnant, 

 and even w!ien this temperature does not generally exceed 

 that of our ulobe *. Now in a granarv the air is not stag- 

 nant, and is never at the same temperature, since it varies 

 annually from — 10° to + 20' (= + 18° to 68° of Fahr.) : 

 xve must therefore seek for the cause of this phoenomenon 

 in other principles than those adopted by the chemists to 

 account for the formation of saltpetre. 



Havinu; reflected on the long continuance of snow on tie 

 foofi of houses, and knowing also that this substance con- 

 tains a considerable quantity of nitric acid, I cannot help 

 thinking that this is the sole cause of the phajuomenon. 

 In short, the tiles under the above circumstances perform 

 the same office with the earths of India and Spain : the part 

 which is inside the granary, being always dry, absorbs the 

 nitrated v.ater f;»uud in its interior; and this water, being 

 dissolved by the anibiciu air, deposits the nitre on the sur- 

 face of this tile, as it deposits it on the surface of the soil 

 of warm countries. Thus in winter it is the nitric acid of 



• l\.e^ nril ile Uttmoires snr le Sulpetrf. 



'Vol. 34. \o. 139. AW. 1809. Y the 



