[ r 75 3 
No kind of air, on which I have yet made the 
experiment, will conduct ele&ricity ; but the colour 
of a fpark is remarkably different in fome different 
kinds of air, which feems to (hew that they are not 
equally good non-conduCtors. In fixed air, the 
eleChic fpark is exceedingly white but in inflam- 
mable air it is of a purple, or red colour. Now, 
ffnce the mod vigorous fparks are always the whitefl:,- 
and, in other cafes, when the fpark is red, there is 
reafon to think that the eieCtric matter paffes with, 
difficulty, and with lefs. rapidity : it is poffible that 
the inflammable air may contain particles which 
conduff electricity, though very imperfectly j and that, 
the whitenefs of the fpark in the fixed air, may be 
owing to its meeting with no conducting particles, 
at all. When an explofion was. made in a quantity 
of inflammable air, it was a little white in the, 
center, but the edges of it were (till tinged with a 
beautiful purple. The degree of whitenefs in this 
cafe was probably owing to the eleCtric matter rufh- 
ing with more violence in an explofion than in a 
common fpark. 
Inflammable air kills animals as fuddenly as fixed 
air, and, as far as can be perceived, in the fame, 
manner, throwing them into convulfions,, and there- 
by occafioning prefent death. I had imagined that,, 
by animals dying in a quantity of inflammable air, 
it would in time become lefs noxious j but this did 
not appear to be the cale ; for I killed a great number 
of mice in a fmall quantity of this air, which I kept 
feveral months for this purpofe, without its being at 
all fenfibly mended ; the laft, as well as the firffi. 
moufe, dying the moment it was put into it. 
3 I once 
