[ 3106 j 
Iwas now thoroushly convinced, 
that the nitrous teit failed entirely 
in fhewing the degree of falubrity 
of this air; for it appeared by this 
method to be nearly dephlogifticated 
air, and yet it was {till a true poi- 
fonous air*, 
Twas indeed | very. forry to find 
thi failure in a method fo well ad- 
: ek forthe exploration of atmo- 
fpheric air. But I am very far 
from thinking that this exception 
dimunifhes in any way the real value 
of the important difcovery, that 
Nitrous air diminifoes refpirable air 
in tbe proportion to its Salubrity. For 
this teit holds good in atmofpheric 
air, which is the chief object of our 
experiments, 
SEC- 
* Does this air owe its a eaplogve nature to the dephlo- 
gifticated air oozing out of the plant? But this very 
air becomes alfo explofive, though it ftands ax ; 
