[ 206 ] 
about eight or nine feet in length, feem to admit as 
large a lateral explofion as any body whatever is 
capable of: for, connecting them with the ground, 
by means of the belt coduCtors (which gave the 
eleCtric matter in the bodies, the freeft recefs pofllble) 
I could never make this explofion much more 
confiderable, ufing the fame jar, and all other cir- 
cumftances the fame. 
It is a manifefl: advantage in thefe experiments, 
that the lateral explofion be not taken from the 
coating of the jar itfelf, or from any part of the 
circuit, very near to it. I have found that, cceteris 
paribus , it is the mod confiderable taken at the ex- 
tremity of a brafs rod, of one foot, or a foot and a 
half long, the other end of which is contiguous to 
the jar. It is analogous to this, that the longed: 
fpark is taken, not from the body oi the prime con- 
ductor itfelf, but at the extremity of a long rod in- 
ferred into it. The eleCtric matter feems to ac- 
quire a kind of impetus by the length of the medium, 
through which it pafles. But I found that the maxi- 
mum, in this cafe, did not exceed, or rather, that it 
did not quite reach, three feet ; for, making ufe of a 
thick iron rod, eight or nine feet long, the lateral 
explofion, taken at the extremity of it, was about 
the fame, as when it was taken at the end of a rod 
four inches from the jar; and not half fo confider- 
able as when taken at the extremity of a rod one 
foot long. This, I imagined, might be owing to 
the obftruClion which the eleCtric fluid meets with 
in palling even through metals ; which appears, by 
my former experiments, to be much more confider- 
able than was generally imagined. 
3 
Upon 
