4 PROCEEDINGS OP THE AMERICAN ACADEMY. 



and eo the quantity of electricity — does not directly apply in its present 

 form. This equation is intended to represent the temperature coefiicient 

 of a complete cell, while we are considering that of a single electrode. 

 In the cell Hg-HgCl-KCl-HgCl-Hg the quantity dQ, which is neglected 

 in the evolution of the above formula, may be an important factor; it is 

 manifest that, since the two ends are alike, this quantity must represent 

 the only difference between the heats of reaction at the hot and cold ends. 



We might nevertheless expect the change of electromotive force with 

 the temperature of a single electrode to be in general affected by the same 

 factors as that of a cell ; in other words, to be independent of every affect- 

 ing cause but the potential, the absolute temperature, and the relation of 

 the change of total energy to the quantity of electricity. "We should 

 expect the nature of the kation in the surrounding solution to have 

 no direct influence, and hence the very large discrepancies observed 

 by Gockel remain without explanation. The matter seemed worth 

 investigation, and two years ago the present work was begun. 



In the first place, many crude experiments were made for the purpose 

 of orientation. These determinations were little more than a repetition 

 of Gockel's, with the addition of a few with more dilate solutions, and 

 the substance of them is presented in the first table below. Each of the 

 figures is the average of a number of experiments, which agreed reason- 

 ably well with one another. Some of these experiments were made by 

 Mr. Harold Edwards, to whom I would here express my thanks. 



SERIES I. 



Preliminary. 



Temperature 20°-55°. 



These results differ materially from Gockel's, which remain incompre- 

 hensible, but the irregularities are still sufficiently marked, especially in 



