ON THE PHYSICAL VIEW OF NATURE. 



121 



extensive class of phenomena which are not produced 

 by mechanical forces, but which result solely from the 

 presence and accumulation of heat. This part of natural 

 philosophy cannot be brought under dynamical theories ; 

 it has principles peculiar to itself, and is based upon 

 a method similar to that of the other exact sciences. 1 

 . . . The dilatations, indeed, caused by the repulsive 

 force of heat, the observation of which dilatations serves 

 as a measure of temperature, are dynamical effects ; but 

 it is not these dilatations which we calculate when we 

 investigate the laws of the propagation of heat." 2 He 

 proceeds to build up this new science " upon a very 

 small number of simple facts, of which the causes are 

 unknown, but which are gathered by observation and 

 confirmed by experiments," 3 and he thus arrives at 

 certain general relations, expressed in the form of equa- 

 tions, which are different from, though analogous to, and 

 not less rigorous than, the general equations of dynamics. 

 One of the great experimental facts upon which Fourier 

 bases his theory of the propagation (i.e., the conduction 

 and radiation) of heat is this, that all motion of heat 

 depends on differences of temperature. He examines 

 how differences of temperature are equalised and de- 

 duces the law of the flow of heat. 4 Although he does 



1 Fourier, ' The"orie analytique,' 

 p. 13. 3 Ibid., p. 14. 



3 Ibid., pp. xi, 18, 39. 



4 I cannot here omit to point out 

 how elegantly Prof. Mach has trans- 

 lated into the language of common- 

 aense the whole process of Fourier 

 for establishing the fundamental 

 equation of the theory. See his 

 ' Principien der Warmelehre ' (Leip- 



zig, 1896), pp. 78, &c., 116 sqq. 

 Every student of physics should 

 read the chapters referring to this 

 subject. The mathematical for- 

 mulae will thus become living to 

 him ; but he will also see how 

 necessary the abstract mathematical 

 expression of common -sense con- 

 ceptions is in order to avoid false 

 reasoning. 



