352 Prof. 0. Lodge on tlie Controversy 



An equation of this general shape is undoubtedly true, 

 but it has not the meaning which they assign to it. 



The II may refer to a particular junction, and in certain 

 cases does, but the E does not. The E means the total 

 E.M.F. of the complete circuit, always. 



I venture to think that this will not be disputed, if the 

 matter be for a few moments carefully attended to. To 

 assist attention we may go carefully over some elementary 

 ground — so elementary that it is difficult to secure for it that 

 amount of attention which is necessary for clearness. 



The thermoelectric power of a metal A at various tempe- 

 ratures means the function which is plotted in an ordinary 

 thermoelectric diagram, with temperatures as abscissae, rela- 

 tively to some standard metal (lead) taken as the line of 

 reference or constant zero of thermoelectric powder. 



To determine this function a circuit has to be constructed 

 with the metal A and the standard metal 0, their junctions 

 brought to different and known temperatures t and t 1 , and 

 the resulting E.M.F. (the whole E.M.F. of the circuit) 

 observed, being measured either by the strength of current 

 developed, or better, by a potentiometer or compensation 

 method. Then the ratio ~Ei/(t' — t) is called the thermo- 

 electric power of the metal A at the mean temperature 

 i(t + t'), and is plotted accordingly, for a considerable range; 

 the differences t' — t being naturally kept small, in case the 

 resulting function is not linear. 



This is done likewise for other metals, B, &c. ; and the 

 relative thermoelectric power of two metals at the same 

 temperature (being Pab = Pb — Pa) is represented by the 

 vertical distance between the A and B lines of the diagram. 



Another function of temperature is likewise represented 

 in the diagram, not by a length but by a rectangular area, 

 bounded on the right by the thermoelectric height, and on 

 the left by the line of absolute zero. This is the function II, 

 such that 



n A=P A T, and n AB = P AB T. 



It is to be observed that in this diagram (apart from the 

 experimental mode of obtaining real values for P) no refer- 

 ence is so far made to any junctions, nor to any circuit, nor 

 current, nor E.M.F., nor evolution of heat ; it represents 

 nothing but a geometrical plotting of certain metallic pro- 

 perties as functions of absolute temperature. 



If we now connect the two metals at one point, they are 

 necessarily at the same temperature, say T 1? at that point ; 



