363. In Chapters VIII and IX we developed a purely mathematical theory, which was found to lead to an explanation of the phenomena of viscosity, conduction of heat and diffusion of gases. This theory, although mathematically perfect, did not go far towards revealing the physical mechanism underlying the phenomena.

There is another method of treating these problems, in which we follow as closely as possible the physical processes which result in the phenomena. This method we now proceed to examine. It does not lead to results possessing the same mathematical exactness as the former method: its importance lies rather in its disclosure of the physical mechanism at work. Briefly, the three phenomena under consideration are regarded as transport phenomena—viscosity is a transport of momentum, conduction of heat is a transport of energy, and diffusion is a transport of mass. The mechanism of transport is provided by the free path; a molecule describing a free path of length λ is in effect transporting certain amounts of momentum, energy and mass through a distance λ. If the gas were in a steady state each such transport would be exactly balanced by an equal and opposite transport in the reverse direction, and the net transport would always be nil. But if the gas is not in a steady state there will always be an unbalanced residue, and this want of balance results in the phenomena we wish to study.