1908] oil the Scientific Work of Lord Kelvin. 217 



Yet if we could reverse all inanimate motion, inorganic nature would 

 unwind again its previous evolution ; and if the materialistic hypo- 

 thesis of life were true, living creatures could grow backwards with 

 conscious knowledge of the future but no memory of the past, and 

 would again become unborn. But the real phenomena of life infinitely 

 transcend human science, and speculation regarding consequences of 

 their ultimate reversal is utterly unprofitable. Far otherwise, how- 

 ever, is it in respect to the reversal of the motions of matter unin- 

 fluenced by life, a very elementary consideration of which leads to the 

 full explanation of this theory of dissipation of energy." * He then 

 follows up the matter by illustrative applications of the theory of 

 probability of a kind that in more recent times has led to a statistical 

 definition of entropy rich in promise for applications to chemistry 

 and to natural radiation. 



The very interesting subject of the thermodynamics of radiation 

 is only about twenty years old. Resting as it does fundamentally on 

 the link with mechanical energy which is afl'orded by Maxwell's work- 

 ing pressure of radiation. Lord Kelvin would never admit its validity. 

 The reason seems to be that he was never able to satisfy himself about 

 any mechanical model of the relation of the atom to the aether that 

 would give a mechanism for this pressural interaction between them. 

 There are those who hold that the physical idea of an electron is 

 sufficiently precise to make the rationale of light-pressure logical and 

 secure. But Lord Kelvin Avould not consider it until he could 

 visualise the whole process — see it in operation, as he used to say — to 

 effect which completely would possibly go deeper than we may ever 

 hope to penetrate ; and this inability cut him off from what some 

 consider to be the most refined and beautiful special development of 

 the science which he founded. 



The question naturally arises, whether the establishment of the 

 mathematical function that is fundamental for the theory of 

 mechanical energy is not a subtler matter than this mere estimation 

 of chances : in other modes of its investigation a powerful array of 

 the dynamical properties of the medium is introduced. What becomes 

 of them in the present aspect ? The answer is, that the chance cannot 

 be estimated aright until we know all the conditions, dynamical and 

 other, to which tbe distribution of molecules is subjected. The 

 dynamical relations find their place as conditions restricting the possi- 

 bilities of random distribution. If through ignorance some of them 

 are overlooked, the chances will be in error ; each new condition that 

 is discovered modifies to some extent the whole process, and thus 

 amends our knowledge. 



But this aspect of entropy is quite in keeping with the subjective 

 character, so to speak, of available energy. Objectively, the dissipa- 



* Cf. Helmholtz's review already quoted, Nature, xxxii., ^1885, Wiss. 

 Abhandlungen, iii. p. 594. 



