n decrease in
the amount of the molecular vibration that constitutes heat.

The question whether there is a limit to the degree of cold possible,
and, if so, where the zero must be placed, was first attacked by the
French physicist, G. Amontons, in 1702-1703, in connexion with his
improvements in the air-thermometer. In his instrument temperatures were
indicated by the height at which a column of mercury was sustained by a
certain mass of air, the volume or "spring" of which of course varied
with the heat to which it was exposed. Amontons therefore argued that
the zero of his thermometer would be that temperature at which the
spring of the air in it was reduced to nothing. On the scale he used the
boiling-point of water was marked at 73 and the melting-point of ice at
51-1/2, so that the zero of his scale was equivalent to about -240 deg.
on the centigrade scale. This remarkably close approximation to the
modern value of -273 deg. for the zero of the air-thermometer was
further improved on by J. H. Lambert (_Pyrometrie_, 1779), who gave the
value -270 deg. and observed that this temperature might be regarded as
absolute cold. Values of this order for the absolute zero were not,
however, universally accepted about this period. Laplace and Lavoisier,
for instance, in their treatise on heat (1780), arrived at values
ranging from 1500 deg. to 3000 deg. below the freezing-point of water,
and thought that in any case it must be at least 600 deg. below, while
John Dalton in his _Chemical Philosophy_ gave ten calculations of this
value, and finally adopted -3000 deg. C. as the natural zero of
temperature. After J. P. Joule had determined the mechanical equivalent
of heat, Lord Kelvin approached the question from an entirely different
point of view, and in 1848 devised a scale of absolute temperature which
was independent of the properties of any particular substance and was
based solely on the fundamental laws of thermodynamics (see HEAT and
THERMODYNAMICS). It followed from the principles on which this scale was
constructed that its zero was placed at -273 deg., at almost precisely
the same point as the zero of the air-thermometer.

In nature the realms of space, on the probable assumption that the
interstellar medium is perfectly transparent and diathermanous, must, as
was pointed out by W. J. Macquorn Rankine, be incapable of acquiring any
temperature, and must therefore be at the absolute zero. That, however,
is not to say