106 THE AIM AND ACHIEVEMENTS OF SCIENTIFIC METHOD. 



have seen that under the guidance of the concept of " heat " as 

 a substance which can be transferred from one body to another 

 it has been found possible to unify the facts of temperature 

 change by the observation that (in an ideal case where only 

 two bodies, A and B, are concerned) changes in temperature of 

 A bear a definite and constant ratio to the like changes of B. 

 The analogy to the law connecting the changes of velocity of 

 mass particles is obvious and might be taken as a support 

 of the view that temperatures may be reduced to dynamical 

 terms ; the state which we call by the name temperature being 

 really a " state of motion " of the particles of a body. But as, 

 following Mr. Eussell and the modern mathematicians, we have 

 seen that velocity is not a state at all, it is clear that the 

 analogy must be differently conceived. The temperature of 

 the body A at a given moment is the " notice " it takes of two 

 preceding pairs of simultaneous temperatures of B and of itself, 

 just as the position of a particle is the " notice " it takes of two 

 preceding configurations of the universe to which it belongs.* 



The only conclusion that it seems legitimate to draw from 

 the consideration oi this and other similar cases is that the law 

 we have been studying is the necessary condition that must be 

 present if it is to be possible to regarfPone kind -of changes in 

 bodies as expressions of the " notice " taken by them of changes 

 of a different order elsewhere. It seems safe to assert that no 

 order of phenomena which does not exhibit this law can be 

 brought into relation with one in which the law obtains, but it 

 would, perhaps, be dangerous to maintain that such orders do 

 not exist. 



* I have ignored in this discussion the fact that the " specific heat " of 

 bodies is only approximately constant. The recognition of this fact would 

 make the argument more complicated, but would not affect it essentially. 

 Cy. J. J. Thomson, Applications of Dynamics to Physics and Chemistry^ 

 1888, p. 90. 



