On the Cause of Heat. 107 
rating power to make up for this loss. A body preserves the 
expansion communicated by heat 7 vacuo, where, insulated from 
all other matter, the only momentum, by which the vibrations of 
its particles can be supported, must have been received before its 
' being thus situated. If we pour mercury into a glass tube shaped 
like a shepherd’s crook, the hook being downwards, the fluid 
will be prevented from occupying that part of the tube where 
the air isin such position as not to escape. In this case, ac- 
cording to the hypothesis in question, the mercury is prevented 
from eutering the space the air occupies, by a series of impalpa- 
ble gyratory movements ; so that the collision of the aérial par- 
ticles against each other, causes each to occupy a larger share 
of space in the manner above illustrated by the descending weight 
and interposed body. The analogy will be greater, if we sup- 
pose a row of interposed bodies alternately striking against each 
other, and the descending weight; or we may imagine a vibra- 
tion in all the particles of the interposed mass equal in aggregate 
extent and force to that of the whole, when performing a com- 
mon movement. If the aggregate extent of the vibration of the 
particles very much exceed that which when performed in mass 
would be necessary to preserve a certain space, it may be sup- 
posed productive of a substance like the air by which the mer- 
cury is resisted. But whence is the momentum adequate in such 
rare media to resist a pressure of a fluid so heavy as mercury, 
which in this case performs a part similar to that of the weight, 
cited for the purpose of illustration? If it be said that the mer- 
cury and glass being at the same temperature as the air, the par- 
ticles of these substances vibrate in a manner to keep up the 
aérial pulsations ; I ask, when the experiment is tried in an ex- 
hausted receiver, what is to supply momentum to the mercury 
and glass? There is no small difficulty in conceiving under the 
most favourable circumstances, that a species of motion, that 
exists according to the hypothesis as the cause of expansion in a 
heated solid, should cause a motion productive of fluidity or va- 
porization, as when by means of a hot iron we convert ice into 
water, and water into vapour. 
How inconceivable is it that the iron boiler of a steam engine 
should give to the particles of water, a motion so totally different 
from any it can itself possess, and at the same time capable of 
such wonderful effects, as are produced by the agency of steam ! 
Is it to be imagined that in particles whose weight does not ex- 
ceed a few ounces, sufficient momentum can be accumulated to 
move as many tons? There appears to me another very serious 
obstacle to this explanation of the nature of heat. How are we 
to account for its radiation im vacuo, which the distinguished 
advocate of the hypothesis has himself shown to ensue? There 
O02 can 
