ON THE PHYSICAL VIEW OF NATURE. 



199 



forced to consider anew the ultimate principles of all 

 physical reasoning, notably the scope and validity of the 

 Newtonian laws of motion and of the conceptions of 

 force and action, of absolute and relative motion, as 

 defined or implied in the mechanical scheme which is 

 based upon them. Also with their increasing com- 

 plexity 1 modern dynamical explanations have undoubt- 

 edly, to every impartial observer, acquired a certain 

 character of artificiality which suggests the question to 

 what extent all such mechanical schemes are an expres- 

 sion of actual truths or merely useful illustrations. For 

 the pursuit of scientific research this question is perhaps 

 of little importance : a method is a correct one if it 

 leads to correct results verified by observation. Philo- 

 sophically, as bearing upon the processes, powers, and 

 limits of human reasoning, the question is all-important. 

 We are thus led beyond the province of scientific into 

 that of philosophic thought. In future chapters we shall 

 frequently have occasion to note this tendency of the 

 purely scientific thought of the century to lead up to 

 philosophical problems. Wherever this is the case a 

 history of scientific thought may legitimately close one 

 of its chapters. 



and physical theory is really the 

 mechanically available energy. . . . 

 This energy is definite, but is not, 

 like matter itself, an entity that is 

 conserved in unchanging amount. 

 ... It may and usually does di- 

 minish, in the course of gradual 

 physical changes." 



* The three volumes of the 

 'Rapports,' &c., mentioned above, 

 have been significantly prefaced by 

 a discourse of M. Poincar^ on the 

 relations of experimental and ma- 

 thematical physics, in which he in- 

 sists upon the unity and simplicity 



eo. 



Artificial 



ei. 



The philo- 



of nature as the two conditions 

 which make generalisations pos- 

 sible and useful. With special ref- 

 erence to modern electrical theories, 

 such as those of Lorentz and Larmor, 

 which he had already criticised in 

 his course on ' Electricite" et Op- 

 tique' (2nd ed., 1901, p. 577, &c.), 

 he discusses the possibility of ulti- 

 mate mechanical explanations. Of 

 these, according to his view, an 

 " infinity " is always possible. He 

 asks what is the aim we are follow- 

 ing " Ce n'est pas le me'canisme, 

 le vrai, le seul but, c'est 1'unite 1 ." 



