274 Prof. Oliver Lodge on the Seat of the Electromotive 



bine, or the junction must cool itself vigorously, — something 

 must disappear at the junction if it is to keep on propelling 

 the current.'" "Not so, my unlearned friend," rejoins the 

 philosopher armed with the elastic hypothesis ; " you forget 

 that it is possible for the specific heat of electricity suddenly 

 to diminish as it flows from copper to zinc ; I have no evidence 

 that it does ; in fact experiment proves that it does not ; but 

 I assert that it must, and in so diminishing it sets free an 

 amount of heat exactly sufficient to maintain the junction at 

 its usual temperature, notwithstanding its energetic propul- 

 sion of the current." " That is very strange/'' muses the 

 other ; " for where metals touch acid, activity is mani- 

 fest 5 and if you change the solution in any way, say by 

 dropping a little bichromate of potash into it, the E.M.F. at 

 once changes ; whereas nothing can seem more passive than 

 the place where the metals touch each other, no chemical 

 action nor any change of temperature goes on there ; you 

 may warm it, or cool it, or hammer it, or put it into all 

 manner of different media, and you cannot change the E.M.F. 

 of the cell an appreciable bit. I cannot see how it is possible 

 that the metallic junction should be propelling the current." 



" No, you wouldn't. You seem to expect some excitement 

 and disappearance of energy wherever work is being done. 

 You seem to expect a kind of Peltier-effect at a junction if it 

 is doing work and propelling the current. You labour under 

 the delusion that wherever energy is transformed from one 

 shape to another, the transformation takes place by means of 

 work being done. I am astonished to find such delusions in 

 the mind of a person so generally sane as yourself.'"' 



This is the view of Ayrton and Perry. If it is not, I 

 challenge them to state it better, and to refute the article of 

 Dr. Hopkinson rather than claim it as on their side. 



Let me restate their assumptions ; every one of them gra- 

 tuitous it seems to me, and unsupported by experiment. 



a. The characteristic function f(t) of a simple thermoelectric 

 circuit is an expression for the Volta-effect between the metals 

 of that circuit. [Mere assumption.] 



b. The Volta-effect of two metals varies with the tempera- 

 ture. [Even this has never been safely proved so as to 

 certainly eliminate all effects due to tarnishing.] 



c. The total E.M.F. of a thermoelectric circuit is equal to 

 the difference between the Volta-effects of its two metals at 

 the temperatures of the hot and the cold junction respectively. 

 [Baseless assumption. Disproved as far as was possible for 

 it to be disproved by the experiments Ayrton and Perry made 

 to support it. See their paper in January, and Prof. Ayrton's 



