82 



SCIENCE. 



[N. S. Vol. XXI. No. 525. 



amount of work clone, at the expense of 

 heat, on unit quantity of electricity while 

 it goes once around the circuit. Evi- 

 dently, then, the area CC'I'IC, which rep- 



FiG. 2. 



resents E, represents also the mechanical 

 equivalent of the net amount of heat con- 

 sumed by unit quantity of electricity in 

 one cycle. 



The arrow-points in Fig. 1 indicate 

 merely the direction of the current result- 

 ing from the net E of the circuit. 



It is consistent with what precedes to 

 consider the area BCC'B'B as representing 

 that part of the total, or net, E which lies 

 in the unequally heated between T and 

 T', the area B'C'l'D'B', as representing 

 that part of E which lies in the junction 

 M^ M., at T', etc. ; and this interpretation 

 is sometimes given as a mere statement of 

 fact. In the course of this paper it will, I 

 hope, be shown that another view of the 

 matter is consistent with the known facts 

 of the case. 



As this declaration puts me for the mo- 

 ment into a somewhat heretical attitude, 

 let me hasten to say that I hold as strongly 

 as any one to the proposition that the area 

 BCC'B'B represents the amount of heat 

 ab.sorbed by unit quantity of electricity in 

 going through the metal from the tem- 

 perature T to the temperature T' , that the 

 area B'C'l'D'B' represents the heat ab- 

 sorbed by unit quantity of electricity in 

 going from il/, to M.^ at temperature T', 

 etc. This propositicm is familiar and needs 

 no proof from irio; but I wish to develop 

 a little one asjject of it which is sometimes 

 overlooked, an aspect which has a decided 

 pedagogic value and which is at least sug- 



gestive of the line of thought I wish to fol- 

 low later. 



As we have in Fig. 1 a diagram in which 

 areas represent heat absorbed, and in which 

 one of the coordinate axes represents tem- 

 perature, the other axis must represent 

 entropy. Let us, therefore, in order to 

 conform to the common practise in the use 

 of the temperatiire-entropy diagram, make 

 the T axis vertical, and the entropy, or S, 

 axis horizontal, thus getting as the equiva- 

 lent of Fig. 1 the Fig. 3. 



(I) /{O 



T 









X' 







— ( — 



\ 





cy 











Fig. 3. 



It is to be observed that Fig. 3 is the 

 obverse of Fig. 1 so that the arrow points, 

 without any relative change of position in 

 going from one figure to the other, now 

 lead clockwise around the area CC'I'IC. 



Any one who is familiar with the tem- 

 perature-entropy diagram of the steam- 

 boiler-engine cycle, as given and discussed 

 by Ewing, will see at once interesting- 

 points of resemblance between Fig. 3 and 

 that diagram. For example, the sloping 

 line CC, which indicates one phase of the 

 Thomson effect, the absorption of heat by 

 the electric current in passing thro\igh the 

 metal il/, from a point at temperature T 

 to a point at the highest temperature, T' , 



