Prof. Challis on a Theory of Molecular Forces. 95 
that the atoms are isolated, and that the sphere of activity of 
attraction is much larger than that of repulsion. These suppo- 
sitions are in accordance with the views now expounded; and the 
explanation there given of the conditions of equilibrium of an 
atom at the boundary applies in the present theory. The prin- 
cipal hypothesis of that explanation is one first admitted by 
Poisson, viz. that, within a distance from the bounding surface 
very small compared to the radius of activity of the molecular 
attraction, there is a rapid increase of density from the surface 
towards the interior. The effect of such change of density will 
be to diminish very much the atomic repulsion on an atom at 
the surface, while the molecular attraction, on account of its far 
greater sphere of activity, will be unaffected by it. The change 
of density must be such that the atomic repulsion at the surface 
is reduced to an equality with the molecular attraction, the latter 
prevailing beyond the surface. 
The conditions of the equilibrium of the atoms situated at and 
near the surfaces of bodies, bring this molecular theory into 
relation with electricity. 
The difference between the circumstances of the equilibrium 
of the superficial atoms of solids and fluids, on which, as said 
above, the difference between the solid and fluid states depends, 
consists, according to these views, in the different amounts of the 
resultant molecular attractions acting in directions parallel to and 
very near the surface, and tending to prevent the separation of 
the atoms in those directions. In fluids, as experience teaches, 
this is a very feeble force; in solids it is overcome by cutting, or 
by fracture, resuming its sway in the new surfaces which these 
operations produce. Atomic arrangement seems to have much 
to do with the energy of this force. 
Both solids and fluids offer great resistance to compression 
within a smaller space. This resistance is due to the atomic 
repulsion, and its energy depends both on the great amount of 
this force, and on its rapid variation with distance. 
It is also a matter of experience that, when the parts of a 
substance (not fluid) are separated, in general they strongly 
resist being joined together again so as to form a single mass. 
This fact may be accounted for if we suppose that the molecular 
attraction which acts on the atoms situated at the boundary of 
the solid, passes through a phase of repulsion before the waves 
to which it is due merge themselves in those that give rise to the 
attraction of gravitation. But independently of such repulsion, 
it is evident that the gradation of density at the boundaries, being 
due to the cause assigned above, must be destroyed before sepa- 
rate portions of the same substance can be perfectly united. In 
eases in which the union is opposed by no energetic molecular 
