Qbfervations on inflammable Air. 339 
time, but without any apparent fign of convullions. 
Thefe experiments, having been often repeated, were 
conftantly attended with the fame event. 
Being allured, beyond any doubt, that the inflam- 
mable air obtained from zinc or iron, and made to 
pafs through quickfilver, was fatal to animals ; I next 
wifhed to oblerve, whether it retained the fame 
properties when it had palled through water; in 
which cafe the volatile fulphurous acid, or other va- 
pour, is abforbed by the water; but, on trying the expe- 
riments, I found that the birds died under thefe circum~ 
fiances as under the others (though not quite fo loon) 
lhewing likewife fome figns of convulfion. I introduced 
feme of this fame air that had pafled through water into 
a giafs tube full of quickfilver, by a method which 
makes the air lofe all its moifture. The birds died in it 
in the fame manner as when the experiment was tried 
upon water. In all thefe cafes the air after the animals 
had died in it was fill inflammable, nor did its exploding 
properties feem to have been at all diminifhed. 
The inflammable air extracted from zinc, and that 
extracted from iron, is fatal to animals even after it has 
been fhaken in water for a minute’s time, or fomething 
longer. By fhaking it a long time, it becomes in fome 
meafure refpirable ; but then it is decompofed in a great 
meafure, 
