[ieee 
fome when melted, but we now know it is the acid that 
ef{capes. 
Tue proportion of ingredients affigned by Mr. -Bergman to 
100 parts of cryftallized nitre is 49 of alkali, 33 of acid, and 
18 of water. He does not feem aware of the lofs occafioned by 
evaporation, nor was he at that time acquainted with the de- 
compofition of the acid occafioned by ignition, both which 
circumftances rendered his conclufions erroneous, as they did my 
own in my firft publication. The lofs occafioned by this latt 
circumftance .he probably attributed to the efcape of water, 
and thence affigned fo large a proportion of this principle. 
According to him it fhould follow that 1oo parts pure alkali 
take up 67,34 of the ftrongeft nitrous acid and afford 204 of 
nitre, yet elfewhere he afferts that 100 parts pure vegetable 
alkali take up but 64 of the ftrongeft nitrous acid. 
Mr. Wenzel’s determination is more exact; he found that 
83,5 grs. of pure vegetable alkali were faturated by 262 of his 
fpirit of nitre, and that the falt thus formed after evaporation 
and expofure to a mild red heat weighed 173.5 grs. and thence 
inferred that 173,5 grs. of ignited nitre contained 83.5 of alkali 
and go of the ftrongeft nitrous acid, confequently 100 parts nitre 
contain about 48 of alkali and 52 of acid. This refult approaches 
much nearer to the truth than that of Mr. Bergman, as it affigns 
a larger proportion of acid than of alkali, which the quantity of 
air expelled from nitre evidently proves, and from this it fhould 
follow that 100 parts of pure alkali take up about 108 grs. of 
the ftrongeft nitrous acid, and afford 208 grs. of nitre, or fome- 
what 
