Obfervations on inflammable Air'. 345 
This experiment fhews, that the air which remained 
in the bladder in the firft experiment was not fo good as 
that breathed thirty-four times fucceffively. In order to 
make this experiment with more precifion, I breathed 
eighty cubic inches of common air, introduced into the 
fame bladder, only eleven times ; beginning after a natu- 
ral expiration. Then I examined this air with the ni- 
trous air, and found that it gave II- 1 3, III + 28. 
Hence it is plain, that the mixture of inflammable and 
pulmonary air breathed eleven times is much inferior to 
common air breathed an equal number of times ; fo that 
there can remain no doubt but that inflammable air is at 
leaft worfe than common air. 
Willing, however, to afeertain this matter flill better, 
I tried to breathe it immediately through a large receiver,, 
partly immerfed in water, and fwimmingin.it, fo that 
the air within the receiver was of the fame elafticity 
with the external air. For this experiment I made ufe 
alfo of a glafs tube bended in two different directions. 
The air contained in the receiver was about 250 cubic 
inches. In all the trials made in this manner, I was ne- 
ver able to breathe the inflammable air more than three 
times, and even at the fecond infpiration I felt a great 
oppreflion. As thefe experiments can be depended upon, 
becaufe they were often and at different times repeated, 
there 
