hygrometer; or, in other words, dense, and containing little else than
the necessary oxygen and azote, and this supplied to a room, fresh and
fresh, in a continual current.'

He next goes on to describe the principle of his new plan of
cooling:--'The method by which I propose to accomplish this
consummation, so devoutly to be desired, is chiefly by taking
advantage of the well-known property of air to rise in temperature on
compression, and to fall on expansion. If air of any temperature, high
or low, be compressed with a certain force, the temperature will rise
above what it was before, in a degree proportioned to the compression.
If the air be allowed immediately to escape from under the pressure,
it will recover its original temperature, because the fall in heat, on
air expanding from a certain pressure, is equal to the rise on its
being compressed to the same; but if, _while the air is in its
compressed state, it be robbed of its acquired heat of compression_,
and then be allowed to escape, it will issue at a temperature as much
below the original one, as it rose above it on compression. Thus the
air, being at 90 degrees, will rise, if compressed to a certain
quantity, to 120 degrees; if it be kept in this compressed and
confined state until all the extra 30 degrees of heat have been
conveyed away by radiation and conduction, and the air be then allowed
to escape, it will be found, on issuing, to be of 60 degrees of
temperature. If a cooler be formed by a pipe under water, and air be
forced in under a given compression at one end, and be made to pass
along to the other, it may thereby, if the cooler be sufficiently
extensive, be robbed of all its heat of compression; and if the
apparatus is so arranged, as it easily may be, that at every stroke of
the pump forcing in air at one end of the pipe, an equivalent quantity
of the cooled compressed air escape from under a loaded valve at the