48 On the Temperatures of Conductors of Electrical Currents. 



required a period of two hours for the temperature of the ther- 

 mometer to coincide with that of the atmosphere. The time 

 necessary for this purpose having elapsed, a current from a 

 single pair was passed in the direction DBA ; the thermometer 

 then rose one degree, and remained stationary there. To break the 

 circuit without disturbing the apparatus, I withdrew the plates 

 of the battery from their cell, when the thermometer quickly 

 rose two degrees above where it had stood while the current was 

 circulating, at which point it remained until it began to cool 

 slowly down by radiation. But without waiting for this, the 

 battery was restored to its cell, which had the effect of speedily 

 reducing the temperature of the joint to within one degree of 

 the surrounding atmosphere. This effect of raising the tempe- 

 rature of a joint two degrees by breaking contact and then 

 depressmg it by the electrical current, admitted of comparatively 

 rapid repetition. I cannot but help viewing it as a beautiful ex- 

 periment, which I should be glad if Messrs. Thomson and Tyndall 

 would accept as a portion of the supposed reduction of tempera- 

 tures by electricity. The elevation of the thermometer CE on 

 breaking contact is what might be expected from the unequal heat- 

 ing effect of the electrical current ; for it is placed at the extremity 

 of the bismuth bar, where the electricity maintains a lower tem- 

 perature than in any other part; when the source of this inequality 

 is removed, it is natural to expect that the heat will diffuse itself 

 equally through all the parts in metallic contact and thus raise 

 the thermometer. On the return of the electrical current the 

 converse of this takes place, the electricity has the power to 

 re-arrange the temperatures after the same manner that it had left 

 them. I do not think that this e;cperiment shows the absorption 

 of heat by electricity, for there is a continual loss of heat going 

 on by radiation from the surface of -the metals ; and if the elec- 

 tricity has the power of concentrating all this loss on one point, 

 it will account for the rapid fall of the thermometer at C. 



To the experimentum crucis given by Dr. Tyndall, the only 

 objection I have to take is the shortness of the time he allows 

 for obtaining the normal temperatures. The late Mr. Troughton, 

 in his account of his method for graduating astronomical 

 circles, speaks of the time required to allow temperatures to 

 settle after handling, and says twelve hours were not too long 

 for his pui-pose. In thermo-electrical inquiries the same patience 

 is necessary, and I have often had two hours to wait for the metals 

 to cool. Should Dr. Tyndall be inclined to do me the honour of 

 again experimenting, I think he may prove that he has been 

 performing one of the same class as that I have met with, 

 where a current of electricity has been shown to be able to re- 

 arrange temperatures, and within certain limits to cool a heated 

 joint at the time it is making the new arrangement. 



