26 
gether untenable; for it is fully decided, both by M. Gay 
Lussac, Mr. Dalton, and others, that all the gases yet 
known, expand equally with equal degrees of heat ; and also 
that their expansion is constantly proportional to their heat. 
We also found it proved, to our conviction at least, that the 
proportions of oxygen and nitrogen gas were nearly the same, 
both in summer and winter, in warm and cold climates. 
These facts of course overthrew all our suppositions to- 
gether, and the question again recurred to us, what can that 
substance or principle be, from which, by the operation of 
the heat, so essential a change is produced in the fitness of 
the atmosphere for combustion ? 
Carbonic acid gas we rejected, its proportion being both 
extremely small, and also subject to the same alteration by 
heat as the other gases. Nor could we connect, in any 
satisfactory way, the agency of electricity or galvanism, which 
is admitted to be of the first importance in some of nature's 
operations. It was therefore more from necessity, than any 
hope of success, that we turned our attention to the aqueous 
vapour existing in the atmosphere.* 
Aqueous vapour, as we have endeavoured to show, does 
exist in the atmosphere to the amount of per cent, in 
weight, or 26 cubic inches per foot in volume. This, it 
must be observed, is at the temperature of 55°, and not 
given as a constant proportional part ; for the force of 
vapour, and consequently its expansion, is variable in differ- 
ent degrees of heat ; nor does this variation follow any 
equable proportion to the accession of heat, as is the case 
with the other gases ; but the degrees of expansion form a 
* The agency of electricity on the atmosphere, we are convinced from 
numerous observations, at different times, is very powerfully operative on blast 
furnaces. But though those effects are quite obvious, they do not appear 
uniform ; nor are we acquainted with any means whereby to ascertain the pecu- 
liar quality or state of electricity in the air, or its respective influence on the 
same, or on the processes of combustion. 
