214 -Dj". Pkiestley’s j Experiments and Obfervations on 
or about 23? grains of nitre cryftallized in mean temperature. 
The fediment of the fame liquor he alfo fuppofed to contain, 
at lead, as much acid as the liquor itfelf. 
That this fediment contains much acid, feems evident from 
this circumftance, that, when it is firft formed, it often emits 
fmall bubbles, which rife to the furface of the liquor, and 
continues to do fo a confiderable time. This was more parti- 
cularly tiie cafe with the fediment which I had from the tinned 
iron tubes. Thefe fmall bubbles, I imagine, confift of nitrous 
air (formed from the fuperabundant acid vapour adhering to 
the metal and the water in the liquor) becaufe when a phial, 
half filled with this liquor, had flood about a week, the air on 
the furface of it inflantly, and repeatedly, extinguifhed apiece 
of lighted wood that was dipped into it. 
From the preceding data, given me by Mr. Keir (and 
making allowance for the indefinite quantity of water con- 
tained in the concentrated acid of vitriol) I am inclined to 
think, that not much more than one-twentieth part of de- 
phlogifticated air is the acidifying principle, and that nineteen 
parts are water. 
This, I would however obferve, relates to air fully fatu- 
rated with water, in confequence of its having been kept in 
jars {landing in water, fo that 1 think it poffible that the water 
in the drieft dephlogifticated air may not amount to more than 
nine- tenths of its weight. But I have not afeertained, by any 
experiment, how much water any of the kinds of air are ca- 
pable of holding in a diffufed Hate, without being any necef- 
farv part of their conffitution. 
Though Mr. Keir found the greatefl part of the acid in 
the liquor with which I furnifhed him to be the nitrous, there 
were evident fig ns of its containing a fmall portion of marine' 
acid, 
