S84 Mr Deuchar on the Nature of Flame. 
Eocp. 20.— A brass ohain was fixed, so as to unite the tube 
with the ground ; under this arrangement, were the flame- elec- 
tric, it must have been conveyed silently by means of the chain 
to the ground ; but this did not take place : the flame still con- 
tinued to dart forward as formerly at each discharge. 
Eocp. 21. — ^When, again, a tube 60 inches in length, was at- 
tached to the apparatus, it was found that the discharges of the 
fulminating powder did not force the flame to the' bottom of 
that tube. Now, had this been the electric fluid, it should have 
passed along any length of tube with equal facility. 
Eaop. 22 A chain was attached to the tube of the apparatus 
when its lei^gth was varied ; and this chain was made to com- 
municate with a Leyden phial. After several trials, duririg 
each of which repeated discharges of the fulminating powder 
were made to pass into the tube, it was found, that there was 
not the slightest indication of the presence of electric fluid in 
the Leyden phial. 
VII. Having found that the first hypothesis does not hold 
true, we are led to adopt the second, which ascribes the pheno- 
mena to condensed caloric. 
In the present case, the caloric may arise from one or more 
of five different sources. 
1. It may be disengaged from the fulminating powder, by a 
change of capacity induced by the blow of the hammer ; or 
probably from a partial decomposition of the union of the sub- 
stances with their natural caloric. 
2. It may arise from the combustion of the ingredients of the 
powder, in contact with the compressed air. 
3. The air of the tube may give out caloric, being condensed 
by the gaseous bodies liberated at the top. 
4. The air in the cap at the top of the tube may give out ca- 
loric, when compressed between the cap in the one direction, 
and the gaseous ingredients of the powder in the other direc- 
tion. And, 
5. The caloric may arise from the rapid movement of the ga- 
seous substances along the brass tube. 
Such are the probable sources from which the caloric may be 
