Eddy.] '^38 [June 16, 



Now, theoretically, no expenditure of energy is necessarry to preserve 

 the uniform velocity of the moving parts of this syren, and once started 

 with a sufficiently high velocity of rotation and proper adjustment of re- 

 ' flectors it would transfer heat from the body A to B regardless of their 

 temperatures, provided no radiations are permitted except those perpen- 

 dicular to the disks, excluding of course all radiations to and from all 

 bodies other than A and B. It would also, as before shown, transfer heat 

 from a colder body to a hotter, even though the radiation follow the 

 general law of radiations from plane surfaces. 



It is needless to state that the action of the syren, regarded as a pos- 

 sible physical process, is directly at variance with the hitherto accepted 

 axioms and conclusions respecting the second law of thermodynamics. 



It is true, we should at first thought be inclined to the belief that the 

 laws of heat should sutler some modification, in case we assume differing 

 rates of propagation not infinite, but we should hardly be prepared to ad- 

 mit the startling conclusions which must flow from such modification, 

 if the physical process just sketched be admitted to be valid, and these I 

 shall now proceed to develop. 



I think it may be readily perceived that the axiom of Clausius, upon 

 which he founds the second law, viz. : that "heat cannot of itself pass 

 from a colder into a hotter body," when applied to radiations, implicitly 

 assumes that the heat is radiated with infinite velocity, for it takes no ac- 

 count of the states of relative rest or motion of the bodies between which 

 heat passes. 



The axiom of Thomson, "it is impossible, by "means of inanimate ma- 

 terial agency, to derive mechanical eft'ect from any portion of matter by 

 cooling it below the temperature of the coldest of surrounding objects," 

 is obnoxious to the same criticism, and, as I have stated elsewhere,* these 

 should not be called axioms at all, since we are not in a position to bring suf- 

 ficient experience to bear upon them to affirm their validity or want of 

 validity. Indeed, if the process of the syren be admitted to be possible, 

 Ave are now in a position to assert that there exists an unexplained con- 

 tradiction, which does not permit us to consider them as applicable to 

 radiations of heat propagated at finite velocities. 



, What, it seems to me, the just quoted statements of Clausius and 

 Thomson really asserted, was the historical fact, that at the date when 

 they were made, no one had as yet invented any machine, or discovered 

 any principle on which it was possible to construct a machine, which could 

 successfully accomplish what these said had not been done ; and it was 

 further implied that no such muchine could probably ever be invented nor 

 any such principle discovered. 



In complete accord with this statement is that of Kirchhofi, made in his 

 lectures upon the Theory of Heat, during the summer semester of 1880, in 

 which he said, if correctly reported, that the second law cannot be (at 



♦Thermodynamics, New York, 1879. 



