PROFESSOR THOMSON ON THE ELECTRO-DYNAMIC QUALITIES OF METALS. 655 
(1) In one or other of the metals (and most probably in both) there must be a 
thermal effect due to the passage of electricity through a non-uniformly heated por- 
tion of it, which must be an absorption of heat or an evolution of heat, according to 
the direction of the current between the hot and cold parts, and proportional in 
amount to the whole quantity of electricity that passes in a stated time. 
(2) The amount of this effect, with the same strength of current and the same 
difference of temperatures, must differ in the two metals to such an extent, that the 
effect of a current in passing from cold to hot in one metal, together with the effect 
of an equal current passing from a place equally hot to a place equally cold in the 
other, may amount to the absorption or evolution, the existence of which has been 
demonstrated. 
15. The reversible thermal effect* of electric currents in single metals of non- 
uniform temperature, which has been thus established, may obviously be called a 
Convection of Heat by electricity in motion. To avoid circumlocution, I shall 
express it that the Vitreous Electricity carries heat with it, when this convection is in 
the “ nominal direction of the current.” On the other hand, when the convection is 
against the “ nominal direction of the current,” it will be said that the Resinous Elec- 
tricity carries heat with it, 
16 to 18. Dynamical Theory applied to draw, from thermo-electric data, inferences 
regarding the Electric Convection of Heat in Copper and in Iron. 
16. The application of the preceding theorem to the particular case of copper and 
iron is a consequence of Cumming’s discovery, that, if one junction in a circuit of 
two arcs of those two metals be kept cold, and the other be heated gradually, a 
current at first sets from copper to iron through the hot junction with increasing 
strength ; but begins to diminish after a certain temperature, which Becquerel 
found to be about 300° Cent., is exceeded ; falls away to nothing when a red heat is 
attained; and sets in the reverse 'f direction when the elevation of temperature is 
pushed higher. 
* See an article by the author, entitled “ On a Universal Tendency in Nature to the Dissipation of Mecha- 
nical Energy” (Proceedings Roy. Soc. Edinb., Feb. 16, 1852, and Phil. Mag., Oct. 1852), where all natural 
operations are divided into two great classes, “reversible” and “irreversible.” See foot-note on § 2 above, 
for an example of the second class. 
t Having myself experienced some difficulty in obtaining the reverse current in the manner described by 
M. Becquerel, in which one junction was heated in the flame of a spirit-lamp, while the other was kept at 
the atmospheric temperature, I found that it could be obtained so as to be observed with the greatest ease by 
means of a very ordinary galvanometer and an iron wire with copper wires twisted round its ends, by keeping 
the lower junction at a temperature considerably above that of the atmosphere, at 100° Cent, for instance ; 
and I ascertained that when both junctions were kept at a very high temperature, in the flame of a spirit-lamp 
for instance, and one of them cooled a little below the temperature of the other, the current produced was the 
reverse of that which the same difference occasioned when both junctions were at ordinary temperatures. See 
Part II. below for further developments on this subject. 
