A fiinity and Freat. 301- 
Nothing of the same kind can be remarked in the-solu- 
tion of liquid, where the contractions follow no simple law, 
and do not enable us to foresee any change in the chemical 
properties of the elements which interpenetrate by solu- 
tion. Further, the thermal phenomena, which may be 
' manifested at the moment in which the liquids interpene- 
trate, do not necessarily imply change of their chemical 
properties. 
Observing that the physical and chemical properties of | 
bodies cannot be separated in a more absolute manner 
than can the phenomena of combination and of solution, 
we conclude, first, that the diffusion of gases is essentially 
different from the solution of liquids, and therefore, that 
whenever, from the mixture of two gases, there results a 
calorific phenomenon, there is a change of condition, and, 
therefore combination. 
The phenomena observed on the contact of liquids and 
solids which mutually dissolve each other is far more com- - 
plex, and deserve a special analysis. 
In attacking this question experimentally, a large num- 
ber of the physical properties of bodies must be known, and 
therefore, be determined whenever they are unknown. 
Hence, all the complications which would be a source of — 
trouble in calculating the effect observed (for instance, the | 
latent heat of the fusion of solids) must be removed at the 
outset in this investigation. Hence, my researches have 
hitherto been limited to the determination of the calorific 
phenomena manifested on the contact of liquids which 
combine or dissolve and produce a liquid. | 
In general, two bodies which dissolve, contract, I shall 
begin by defining what I call heat of contraction, either in 
the particular case of liquids, or in the general case. OTR 
Suppose we take a body whose weight is.unity, knowing ° 
the law of its expansion as a function of the temperature, 
we can calculate the temperature at which this body would 
lose a given fraction of its volume; and knowing the specific 
heat of this body within the limits of experiment, we can — 
calculate the heat of contraction corresponding to the dimi- 
nution of volume. Hence we can obtain the quantity of 
heat necessary for a given variation of the density. That 
will be the heat of contraction. 
Suppose we take water and sulphuric acid at O? super- 
posed in a spherical flask provided with a perfectly cylin- 
drical narrow neck; suppose that the two surfaces of 
contact are separated by an obstacle easy to break—such, 
