82 Relations between Chemical Vhange, [January, 



and make the total resistance the same, either by shortening 

 the conducting wire, or by enlarging the zinc and platinum 

 plates, or in any other manner. We now get, as before, 

 exactly double the consumption of zinc, but the quantity of 

 electricity circulating and the heat evolved in the circuit 

 (if the accepted formulae be right) the same. 



How is this? Is it again that the heat evolved by a 

 given weight of zinc is only half under this arrangement, or 

 is only half of it put into circulation ? In either case the 

 ordinarily accepted principles are hopelessly wrong. 



It is obvious that we might in a similar manner, by the 

 use of zinc with different metals as a negative, multiply to 

 any extent the cases to which the formulae will not apply. 

 Let us take one more of a different kind. 



9. But before describing it let me observe that, in speaking 

 of the heat produced by the oxidation of a given weight of 

 zinc, we speak of what is left of this heat after deducting 

 the amount of cold produced by the evolution of the corres- 

 ponding amount of hydrogen at the opposite pole of each cell. 

 On the principles ordinarily accepted, if this credit balance 

 of heat were nil, the electricity and heat evolved in the 

 circuit should be nil, and if it were a minus quantity, cold 

 instead of heat should be produced in the circuit. 



But now take a couple consisting of copper and platinum. 

 The heat produced by the oxidation of an equivalent of 

 copper is said to be 21,885 units, but the cold produced by 

 the evolution of an equivalent of hydrogen 34,462 units. 

 Hence, if the amount of heat evolved in a circuit be equal to 

 that produced by the oxidation of the metal, minus the cold 

 evolved by the hydrogen, the wire should be cooled and not 

 heated ; and yet copper is universally recognised as positive 

 to platinum. Or take again an alloy of zinc and copper or 

 other metals, whose, heat-equivalent is the same as that of 

 hydrogen, or 34,462. There can be no doubt that this 

 would be positive to platinum, and would produce a current 

 of electricity, though the heat evolved would be nil ; and we 

 should have the anomaly of an electric current passing 

 through a homogeneous wire without heating it. 



10. We conclude, therefore, that the whole subject 

 requires a fresh, strict, and full experimental investiga- 

 tion^ What we want to discover is how much heat 

 a certain consumption of zinc and other metals produces 

 when used in different electro-motive combinations with 

 other metals, and what becomes of it ; how much circulates 

 through the circuit, and according to what laws. An 

 investigation of this kind would probably show either that a 



