67 
under great pressure, nearly the same as fire is to gunpowder. 
And this will account for the most destructive cases of boiler 
explosions ; whilst those of a more harmless nature show that 
the fractures were small at first, and then gradually extended. 
He also objected to the term “ superheated steam,” as being 
inapplicable to it in any state; because, when steam is in 
contact with water, it will be of the same temperature as the 
water; and if heated apart from water, the same laws of 
expansion by heat apply to steam as to air, and neither can be 
“ superheated,” though made very hot. 
Again, steam can never be “mixed up with the water” in 
a boiler when both are under the same statical pressure, and 
the steam formed will rise into the chamber, so that the water 
will always be in contact with the boiler except when steam 
is drawn off. Still, in rapid escapes, it may drive out water 
and become entangled therewith, as in many explosions. 
It having been shown that most, if not all, explosions are 
occasioned by simple steam pressure, acting on the weakest 
parts at first, and thence extending more or less rapidly, it 
would seem needless to seek for any other cause or force to 
account for them ; yet, in some cases, the effects appear to 
imply a more sudden and violent action, like that of explosive 
compounds. In such instances, may they not arise from the 
actual decomposition of the water by heat alone ? Although 
we have high authority (cited) against this, yet the Author 
held it rash to conclude that water could not be resolved into 
its constituent gases by direct action of heat from the boiler 
upon water pressed into contact with the metal plates. It 
has been proved, long since, that by heat, in the most intense 
form known to us — that of electricity — water is decomposed 
and both of its constituent gases are liberated. Therefore, 
since no evidence has been adduced to show that this does not 
take place in any water when so confined and heated, the 
affirmative may at least be possible, and seems probable, in 
some instances, as before named. 
However, he held it desirable that the question should, if 
possible, be set at rest by experiment; and to this end a 
method was suggested for putting the matter to a direct test ; 
