Chemical Affinity in terms of Electromotive Force. 73 
the expense of the sensible heat of the cell, which becomes 
cooled by the passage of a current of too small magnitude to 
generate, in accordance with Joule's law, sufficient heat in the 
cell to overpower this cooling action. Inasmuch, however, as 
the mercury-calorimeter was employed in these experiments 
of Favre, whilst, from the nature of the case, but feeble cur- 
rents passed, so that the total amount of chemical action in a 
given time could be but small, it seems not unlikely that an 
excessively large probable error attends the numerical values 
obtained. In point of fact, one of the cells found by Favre 
to behave in this way was Grove's cell; and his results in this 
repect are totally at variance with all other experiments on 
the subject (compare H. F. Weber, Phil. Mag. 1878, v. p. 195), 
leading to the conclusion that the supposed cooling action was 
not a real effect, but simply the result of the accumulation of 
experimental errors. In order to see whether this was also the 
case with the other cells examined by Favre, the following 
experiments were made. 
These other cells were simple voltaic couples of zinc and 
platinum or cadmium and platinum immersed in dilute hydro- 
chloric acid ; the numbers obtained by Favre as the cooling 
effects per gramme equivalent of metal dissolved were respec- 
tively 1051 and 1288 gramme-degrees, corresponding with 
•046 and *057 volt*. On the other hand, with dilute sulphuric 
acid in lieu of hydrochloric, Favre found that no cooling action 
was traceable, but that the cells were always warmed by the 
passage of a current. Now these results, if correct, must 
imply that the E.M.F. of a zinc-platinum or a cadmium- 
platinum cell, when generating only a minute current, is above 
the value corresponding with the heat-development due to the 
net chemical action taking place when hydrochloric-acid solu- 
tion is the exciting fluid, and be^w that value when dilute 
sulphuric acid is used instead; i. e. the electromotive forces of 
cells containing dilute hydrochloric acid must be above '754 
and *388 volt respectively with zinc-platinum and cadmium- 
platinum couples, and the electromotive forces of cells con- 
taining dilute sulphuric acid must be below *835 and '470 volt 
respectively with these same couples, these being the values 
in E.M.F. corresponding respectively with the heat-develop- 
ments per gramme-equivalent in the reactions 
* For the sake of comparison r with the experiments described in the 
previous portions of these researches, the factor 4410 for converting 
gramme-degrees into volts is adhered to, notwithstanding that the balance 
of evidence now seems to indicate that the value of J hitherto assumed 
(42 megalergs) is somewhat too high, and that the B.A. unit of resistance 
is upwards of 1 per cent, below its intended value, instead of being exact 
as hitherto assumed. 
