372 Lord Kelvin [April 27, 



equilibrium with its constituents in any state of relative motion, no 

 atom will fly away from it, provided the total kinetic energy of the 

 given initial motion does not exceed some definite limit. A gas is a 

 vast assemblage of molecules thus defined, each moving freely through 

 space, except when in collision with another cluster, and each retain- 

 in^ all its own constituents unaltered, or only altered by interchange 

 of similar atoms between two clusters in collision. 



§ 20. For simplicity we may suppose that each atom, A, has a 

 definite radius of activity, a, and that atoms of different kinds, A, A', 

 have different radii of activity, a, a' ; such that A exercises no force on 

 any other atom, A', A", when the distance between their centres is 

 greater than a + a' or a + «"• We need not perplex our minds with 

 the inconceivable idea of " virtue," whether for force or for inertia, 

 residing in a mathematical point * the centre of the atom ; and with- 

 out mental strain we can distinctly believe that the substance (the 

 " substratum " of qualities) resides, not in a point, nor vaguely 

 through all space, but definitely in the spherical volume of space 

 bounded by the spherical surface whose radius is the radius of 

 activity of the atom, and whose centre is the centre of the atom. In 

 our intermolecular forces thus defined, we have no violation of the old 

 scholastic law, " Matter cannot act where it is not," but we explicitly 

 violate the other scholastic law, " Two portions of matter cannot simul- 

 taneously occupy the same space." We leave to gravitation, and 

 possibly to electricity (probably not to magnetism), the at present 

 very unpopular idea of action at a distance. 



§ 21. We need not now (as in § 16, when we wished to keep as 

 near as we could to the old idea of colliding elastic globes) suppose 

 the mutual force to become infinite repulsion before the centres of two 

 atoms, approaching one another, meet. Following Boscovich, we 

 may assume the force to vary according to any law of alternate 

 attraction and repulsion, but without supposing any infinitely great 

 force, whether of repulsion or attraction, at any particular distance ; 

 but we must assume the force to be zero when the centres are coin- 

 cident. We may even admit the idea of the centres being absolutely 

 coincident, in at all events some cases of a chemical combination of 

 two or more atoms ; although we might consider it more probable 

 that in most cases the chemical combination is a cluster, in which the 

 volumes of the constituent atoms overlap without any two centres 

 absolutely coinciding. 



§22. The word "collision" used without definition in §19 may 

 now, in virtue of §§ 20, 21, be unambiguously defined thus : 

 Two atoms are said to be in collision during all the time their 

 volumes overlap after coming into contact. They necessarily in 

 virtue of inertia separate again, unless some third body intervenes 

 with action which causes them to remain overlapping ; that is to say, 



* See Math, and Phys. Papers, vol. iii. art. xcvn. ' Molecular Constitution 

 of Matter,' § 14. 



