054 PROFESSOR THOMSON ON THE ELECTRO-DYNAMIC QUALITIES OF METALS. 
the neutral point, cannot, in the present state of science as regards the theory of heat, 
be reasonably doubted. 
12. If, therefore, a circuit of two metals have one junction kept at the neutral 
point, and the other at some lower temperature, the current excited will cause the 
evolution of heat at the cold Junction, but neither absorption nor evolution of heat 
at the hot junction; and in the rest of the circuit there will be effects either purely 
thermal, or thermal and mechanical or chemical, according to the nature of the 
resistance against which the electro-motive force is allowed to work. The source 
from which the electro-motive force derives its energy to produce these effects cannot 
be at the hot junction (§ 10), where heat is neither absorbed nor evolved, nor at the 
cold junction (§ 11), where heat is evolved, nor of course in any uniformly heated 
part in either metal, through all of which, provided the metal has no thermo-electric 
crystalline characteristic, there can be nothing but a frictional evolution of heat; 
that is, it is nowhere but in those portions of the circuit where the temperature 
varies between that of the cold and that of the hot junction. In those portions, 
therefore, there must be as much heat absorbed, in virtue of the current, as is equi- 
valent to the aggregate mechanical value of the heat evolved at the cold junction, 
and all the effects, thermal, mechanical, and chemical, produced in the rest of the 
circuit. 
13. If, for example, an electro-magnetic engine be introduced into the circuit, and 
be allowed to work at such a rate as to reduce, by its inductive reaction, the strength 
of the thermo-electric current to an infinitely small fraction of what it is when the 
engine is at rest, the heat absorbed in virtue of the current in the unequally heated 
parts of the two metals will be equal to the heat evolved at the cold junction, together 
with the thermal equivalent of the work done by the engine, and will be simply pro- 
portional to the strength of the current. On the other hand, if the etigine be forced 
to work a little faster, so as to overbalance by an infinitely small amount the ther- 
mal electro-motive foree, and cause a reverse current in the circuit, there must be 
heat evolved in virtue of this current in the unequally heated parts of the two metals 
to an amount equal to the heat absorbed at the cold junction, together with the ther- 
mal equivalent to the work done against electro-magnetic forces in the engine. It 
follows that in the unequally heated portions of the two metals, the current passing 
from cold to hot in one, and from hot to cold in the other, must produce a thermal 
effect, in simple proportion to its owm strength, constituting on the whole an absorp- 
tion of heat when the thermal electro-motive force is allowed to produce a current, 
and an evolution of heat when a current is forced by other means in the contrary 
direction. 
14. Hence, for any two metals which are thermo-electrically neutral to one another 
at a certain temperature, and which possess reverse thermo-electric properties for 
temperatures above and below the neutral point, we conclude the following propo- 
sitions ; — 
