360 Prof. E. Edlund on the New Electromotive 



height as that of the positive. If, also, the temperature of the 

 negative pole be only so much raised that the difference between 

 the poles is decreased, the thermoelectric current must be dimi- 

 nished; or it might even be possible that it would sink to zero, and 

 then acquire an opposite direction, before the temperature of the 

 negative pole attained that of the positive. If the statement that 

 the current decreases when the temperature of the negative pole 

 is increased is perfectly correct under all circumstances, the arc 

 must of course retain its conductivity unchanged. Yet this de- 

 pends principally, if not exclusively, on the quantity and con- 

 dition of the transported particles, and not on the temperature of 

 the gas. On the contrary, it is by no means certain a priori that 

 the current will increase if the temperature of the positive pole be 

 raised, and thereby the difference between the temperature of the 

 poles become greater, because it is possible that the greatest 

 electromotive force corresponded to a certain temperature of the 

 positive pole, and that the force would become less at a higher 

 temperature. It is also important to be able to show that in- 

 crease in temperature in the negative pole produces no diminu- 

 tion in the intensity of the currrent. 



To effect this, a Bunsen's burner provided with a bellows was 

 used. When the flame was directed against carbon of the same 

 quality as that used in the experiments, the carbon began to ig- 

 nite, and a thin wire coiled round it melted. The temperature, 

 at any rate on the side of the carbon turned towards the flame, 

 was 1000° or more. Upon this the following experiments were 

 made, partly without heating by the burner, partly with the ap- 

 plication of heat to the positive or negative pole as near as 

 possible to the arc. 



Experiment 7. The battery consisted of fifty elements; but the 

 current, probably owing to imperfect conduction, was not parti- 

 cularly strong. The magnetometer was connected with the 

 principal circuit as shown in fig. 1 ; that is, the electromotive 

 force in the luminous arc caused the deflection. 



The mean of ten observations, when there was no heating, and 

 when the negative pole was heated to strong ignition, gave : — 



Without heating .... 30*8 divisions. 

 The probable error . . . . +2 '5. 



Heating the negative pole . 40*8 divisions. 

 The probable error . . . . +3*1. 



On heating the negative pole, therefore, and diminishing thus 

 the difference in temperature between the poles, the current was 

 stronger — quite opposite to what should have been the case if the 

 current had been of thermoelectric origin. In the subsequent 



