Prof. Thomson on the Dynamical Theory of Heat. 219 



where W denotes the aggregate amount of work spent in pro- 

 ducing the operations, and J the mechanical equivalent of the 

 thermal unit. 



§§ 102-106. Initial examination of Thermo-dynamic circum- 

 stances regarding Electric Currents in Linear Conductors. 



102. Peltier's admirable discovery, that an electric current in 

 a metallic circuit of antimony and bismuth produces cold where 

 it passes from bismuth to antimony, and heat where it passes 

 from antimony to bismuth, shows how an evolution of mecha- 

 nical effect, by means of thermo-electric currents, involves 

 transference of heat from a body at a higher temperature to a 

 body at a lower temperature, and how a reverse thermal 

 effect may be produced, by thermo-electric means, from the ex- 

 penditure of work. For if a galvanic engine be kept in motion 

 doing work, by a thermo-electric battery of bismuth and anti- 

 mony, the current by means of which this is effected passing, as 

 it does, from bismuth to antimony through the hot junctions, 

 and from antimony to bismuth through the cold junctions, must 

 cause absorption of heat in each of the former, and evolution of 

 heat in each of the latter; and to sustain the difference of tem- 

 perature required for the excitation of the electromotive force, 

 even were there no propagation of heat by conduction through 

 the battery, it would be necessary continually, during the exist- 

 ence of the current, to supply heat from a source to the hot 

 junctions, and to draw off heat from the cold junctions by a 

 refrigerator : — Or, if work be spent to turn the engine faster 

 than the rate at which its inductive reaction balances the elec- 

 tromotive force of the battery, there will be a reverse current 

 sent through the circuit, producing absorption of heat at the 

 cold junctions, and evolution of heat at the hot junctions, and 

 consequently effecting the transference of some heat from the 

 refrigerator to the source. 



103. We see, then, that in Peltier's phsenomenon we have a 

 reversible thermal agency of exactly the kind supposed in the 

 second law of the dynamical theory of heat. Before, however, 

 we can apply either this or the first law, we must consider other 

 thermal actions which are involved in the circumstances of a 

 thermo-electric current ; and with reference to the second law, 

 we shall have to examine whether there are any such of an 

 essentially irreversible kind. 



lot. It is to be remarked, in the first place, that a current 

 cannot pass through a homogeneous conductor without genera- 

 ting heat in overcoming resistance. This effect, which we shall 

 call the frictional generation of heat, has been discovered by 

 Joule to be produced at a rate proportional to the square of the 



Q2 



