It is, perhaps, more natural to suppose, that the particles of caloric have a stronger mutual attraction than those of any other substance, and that these latter particles are forced asunder in consequence of this superior attraction between the particles of the caloric, which forces them between the particles of other bodies, that they may be able to reunite with each other. We have somewhat analogous to this idea in the phenomena which occur when a dry sponge is dipt into water: The sponge swells; its particles separate from each other; and all its intervals are filled up by the water. It is evident, that the sponge, in the act of swelling, has acquired a greater capacity for containing water than it had when dry. But we cannot certainly maintain, that the introduction of water between the particles of the sponge has endowed them with a repulsive power, which tends to separate them from each other; on the contrary, the whole phenomena are produced by means of attractive powers; and these are, _first_, The gravity of the water, and the power which it exerts on every side, in common with all other fluids; _2dly_, The force of attraction which takes place between the particles of the water, causing them to unite together; _3dly_, The mutual attraction of the particles of the sponge with each other; and, _lastly_, The reciprocal attraction which exists between the particles of the sponge and those of the water. It is easy to understand, that the explanation of this fact depends upon properly appreciating the intensity of, and connection between, these several powers. It is probable, that the separation of the particles of bodies, occasioned by caloric, depends in a similar manner upon a certain combination of different attractive powers, which, in conformity with the imperfection of our knowledge, we endeavour to express by saying, that caloric communicates a power of repulsion to the particles of bodies.
[2] Whenever the degree of heat occurs in this work, it is stated by the author according to Reaumur's scale. The degrees within brackets are the correspondent degrees of Fahrenheit's scale, added by the translator. E.
[3] Collections of the French Academy of Sciences for that year, p. 420.