NOVEMBKR 20. 1903.] 



SCIENCE. 



643 



tion, rotatory motion, motion involved in 

 simple types of elastic distortion, steady 

 types of fluid motion including: wave mo- 

 tion and the corresponding types of force 

 action. This definition of mechanics in- 

 cludes a large portion of the subjects of 

 electricity and magnetism and of light. 

 Mechanics, as here defined, includes all 

 phenomena which involve action between 

 bodies and regions of finite size, which ac- 

 tion can be perceived directly or indirectl.y 

 as a unit. 



Thermodynamics treats of those phe- 

 nomena which can be to a high degree of 

 approximation correlated in one-to-one cor- 

 respondences to states and processes in- 

 volving thermal equilibrium and involving 

 those permanently varying states which I 

 have called steady sweeps. Thermody- 

 namics involves much mechanics, but me- 

 chanics proper ignores everything which 

 pertains strictly to thermodynamics. 

 Thermodynamics includes all portions of 

 the subjects of electricity and magnetism 

 and light which are not completely defined 

 in mechanical terms. 



Statistical physics is the study of all the 

 actual physical phenomena of natvire, some 

 of which, indeed, may be described in terms 

 of the notions of ideal mechanics to a high 

 degree of approximation, and some of 

 which, indeed, may be described in terms 

 of the notions of ideal thermodynamics to 

 a high degree of approximation, but all of 

 which are more or less erratic and in their 

 minute details infinitely manifold, and in 

 all of which the notion of one-to-one cor- 

 respondence or of cause and effect, if one 

 prefers that mode of expression, fails. 



A clear understanding of the essential 

 limitations of .systematic physics is im- 

 portant to the engineer; it is. I think, 

 equally important to the biologist, and it 

 is of vital importance to the physicist, for 

 in the ease of the physicist, to raise the 

 question as to limitations is to raise the 



question as to whether his science does 

 after all deal with realities, and the con- 

 clusion which nuist force itself on his mind 

 is, I think, that his science, the systematic 

 part of it, comes very near, indeed, to 

 being a science of unrealities. This is not 

 necessarily to the discredit of the physicist, 

 provided he Icnows it. 



The engineer is not far wrong in his 

 application of the principles of mechanics, 

 for the engineer is chiefly concerned with 

 integral relationships between finite things. 

 Neverthele.ss, I think that the engineer fre- 

 quently attempts to carry his mechanics 

 too far, when, for example, he attempts 

 anything but the crudest correlation in his 

 studies of such things as friction and fluid 

 motion. The phenomena of friction and 

 of fluid motion can not be correlated— I 

 do not mean by human means, condition- 

 ally as it were, but I mean that they abso- 

 lutely can not be approximately correlated 

 by any means in one-to-one correspond- 

 ences. I discussed this matter briefly in 

 some remarks before the American Insti- 

 tute of Electrical Engineers on December 

 19, 1902.* In studies which require the 

 application of the principles of thermo- 

 dynamics, on the other hand, engineers are, 

 I think, frequently in error. Thus, Mr. 

 Swinburne's difficulty— his statements, be- 

 ing partly right and partly wrong, may 

 be taken to indicate a difficulty— seems to 

 me to lie in an improper application of 

 the principles of thermodynamics. 



The biologist, on the other hand, is, I 

 think, usually illogical when he attempts 

 to make use of the ideas of systematic 

 physics. The biological sciences, in so far 

 as they are related to systematic physics 

 at all, are related primarily to thermo- 

 dynamics, and in so far as the biologist is 

 unfamiliar with the principles of thermo- 

 dynamics he can not make proper use of 



* Soe Tiaiis. .1. /. E. E., January. 1!)03. pp. 

 70-80. 



