t 94 ] 
Caine, which is hard, but dear and fweet, a candle 
will not burn in it after three minutes. The fame is 
the cafe with the rain water, which I get from 
the roof of my houfe. But in diddled water, or 
the water of a fpring-well near the houfe, I muft 
agitate the air about 20 minutes, before it will be 
fo much injured. It may be worth while, to make 
farther experiments, with refpect to this property 
of water. 
In confequence of ufing the rain water, and the 
Well water above-mentioned, I was very near con- 
cluding, contrary to what 1 have aflerted in my 
printed papers, that common air differs a decom- 
pofition by great rarefaflion. For when I had 
collected a confiderable quantity of air, which had 
been rarefied about four hundred times, by an ex- 
cellent pump made for me by Mr. smeaton, I al- 
ways found, that when I filled my receivers with 
the water above-mentioned, though I did it fo 
gradually as to occafion as little agitation as pofiible, 
a candle would not burn in the air that remained 
in them. But when I ufed diddled water, or frefh 
ipring water, I uudeceived mylelf. 
I think mvfelf honoured by the attention, which, 
from the fird, you have given to my experiments, 
and am, with the greated refpeft, 
Dear Sir, 
Your mod obliged 
humble Servant, 
London, 7 Dec. 1773. 
j. PRIESTLEY. 
POST- 
