1 



67 

 1 



ill 



34 9 



Intelligence and Miscellaneous Articles. 395 



certain current, taken as unity, in passing- from copper to the metals 

 afterwards designated. By comparing the number of thermal units 

 found with the thermic equivalent of the chemical effects produced 

 by a current of the same intensity in a sulphate-of-copper element 

 (in which voltaic and chemical heat appear to be exactly equal), I 

 can compare with the electromotive force of this element the elec- 

 tromotive forces of the Pelterian kind which exist at the junction of 

 the copper with the metal after it. I thus find that at a tempera- 

 ture of 25° these electromotive forces are represented by the follow- 

 ing fractions : — 



Copper-antimony and cadmium alloy of M. Becquerel 



Copper-antimony 



Copper-iron 



Copper-zinc 22 1 71 



Copper-cadmium tqit 



Copper-brass ^g- 



Copper-bismuth -^ 



Copper-bismuth and antimony of M. E. Becquerel. ^ 

 I then seek to know if, for the couple copper-bismuth of M. E. 

 Becquerel, the variation of this force between two temperatures, 25° 

 and 100°, can account for the electromotive force between the same 

 limits of temperature, a force which M. Becquerel estimated by 

 taking as unit a sulphate-of-copper couple. With this view I measure 

 in an appropriate stove the Pelterian effects at the two temperatures 

 indicated ; I find between the result predicted and that given by ex- 

 periment a slight difference. But it is not the less constant that in 

 this couple the electromotive forces of the Pelterian kind are much 

 predominant. 



7. The seventh paragraph is devoted to the investigation and mea- 

 surement of the Thomsonian effect. 



I have commenced by verifying that the Thomsonian effect was 

 proportional to the intensity of the current. 



The effect in question may be altered by several disturbing causes 

 — want of homogeneity, tempering, twisting, crj^stalline texture, 

 &c. These are effects of the Pelterian kind ; they are proportional 

 to the intensity of the current, but they change the sign when, other 

 things being equal, end for end the conductors are changed ; hence 

 this gives a method of eliminating disturbing causes by two opera- 

 tions made on the same conductors reversed. 



8. In the eighth paragraph I attempt to value the relative part of 

 the electromotive forces of the Pelterian and of the Thomsonian kind. 

 From the point of view of the direction, it is found that in Becque- 

 rel's copper-bismuth couple and in the copper-iron couple (before inver- 

 sion) these two kinds of electromotive force add on to one another. 



9. From what precedes, leaving aside the Thomsonian effects, it 

 may be regarded as an experimental law that thermoelectric couples 

 are proportional for all couples, between the same temperatures, to 

 the electromotive forces which have their seat at the surfaces of junc- 

 tion. In other words, the electromotive force is for each a function 

 of the temperature ; the ratio of a finite variation of the value of this 

 function to the value of this function for determinate temperatures is 

 the same for all couples, which requires that this function of the 

 temperature be the same for all except for a very small factor. 



