[ 524 ] 



LXIIL A Kinetic Theory of Solids, with an Experimental 

 Introduction. By William Sutherland. 



[Concluded from p. 225.] 

 Theory. 



THE working out of a kinetic theory of solids ought, on 

 purely theoretic grounds, to have proved a simpler matter 

 than that of the kinetic theory of gases, because in solids 

 each molecule abides about a fixed mean position ; but the 

 generality of the two great experimental laws of perfect 

 gases stimulated theory to overcome the more difficult task 

 first. The experimental laws which have just been estab- 

 lished for solids seem to give as good a guarantee that in 

 its essence the kinetic theory of solids must be as simple a 

 matter as that of gases. 



We will therefore start out on the assumption that in solids 

 the molecules have the properties assigned to them in the 

 kinetic theory of gases, that they attract one another, and 

 that each one moves in a small region of space round a 

 certain mean position. In the almost purely statical mole- 

 cular theories of solids put forth in the early part of this 

 century by the great French elasticians, repulsive forces had 

 to be imagined to equilibrate the attractive ; for example, 

 Poisson's molecular theory was founded on the supposition 

 that while molecules of matter attracted molecules of matter, 

 they repelled the particles of heat, so that each molecule was 

 in equilibrium under two sets of opposite central forces. But 

 it is easy to see that no theory founded on purely central 

 attractions and repulsions can explain the facts of rigidity as 

 we actually know them, because in a pure shear, as there is 

 no change of volume, there can be no change in the potential 

 energy of the purely centrally acting molecules ; or, more 

 accurately, a small shear of the first order will produce a 

 change of potential energy of the second order of small 

 quantities, but experiment shows that the change of energy 

 in a shear is of the same order of magnitude as the change 

 that accompanies a change of volume : hence purely central 

 attractions and repulsions, while they can give a sort of 

 rigidity, give one infinitely different from that of nature. 

 The statical theory can give equations of the same form as 

 the natural ones, but with some of the coefficients infinitely 

 smaller. 



Attempts have been made to avoid this difficulty by giving 

 the molecules polar properties, which, however, can hardly be 



