TOLMAN. — ELECTROMOTIVE FORCE PRODUCED IN SOLUTIONS. Ill 



able to detect the existence of potential differences between the upper 

 s,nd lower electrodes placed in solutions of silver nitrate and cadmium 

 iodide. The small electromotive forces produced in these solutions by 

 gravity are masked by the presence of much larger variable potential 

 differences which are due to otherwise undetectable differences between 

 the two electrodes used. The effect of gravity is only perceived when 

 a series of measurements is made of the deflections of a very sensitive 

 galvanometer, the positions of the upper and lower electrodes being 

 interchanged between each two measurements by a reversal of the tube 

 containing the electrolyte. The potential differences, which Colley 

 thus found, were of the order of magnitude theoretically predicted. 

 Later he also showed the presence of a momentary electromotive force 

 produced by the sudden stopping of a falling tube containing cadmium 

 iodide solution. 



The experimental problem was next attacked by Des Coudres who 

 substituted centrifugal force for the weaker action of gravity. He 

 made use of a rotating apparatus in which were placed two tubes with 

 electrodes connected in series, containing cadmium iodide solution. The 

 electrodes were placed 31 cm. and 9 cm. from the center and at 5.8 

 revolutions per second he obtained an electromotive force of 155 micro- 

 volts. Individual measurements, however, varied at times more than 

 10% from the mean. Although the substitution of centrifugal force for 

 the force of gravity greatly increases the size of the effect which can be 

 obtained, it is accompanied by the introduction of new errors, produced, 

 for example, by the unequal heating of the central and peripheral por- 

 tions of the apparatus, as well as by other difficulties. The method was 

 abandoned by Des Coudres in favor of measurements on specially con- 

 structed gravity cells. These measurements are of a considerable 

 degree of accuracy and the results will be given in detail. 



As we shall see in the next section, it is possible to derive a relation 

 connecting the electromotive force produced by the action of gravity 

 and the transference number of the electrolyte. The measurements of 

 Des Coudres were made for the purpose of comparing the transference 



Chem. 5, 157 (1890); Bredig Zeitschr. f. phys. Chera. 17,459 (1895); Lobry 

 de Bruyn et van Calcar, Red. d. travaux chim. d. Pays-Bas, 23, 218 (1904); 

 Franklin & Freudenberger, Trans. Am. Electrochem. Soc. 8, 29 (1905); 

 Earl of Berkeley & C. V. Burton. Phil. Mag. 17, 606 (1909); Gibbs, The 

 Scientific Papers of, Longmans, Green & Co. (1906), Vol. I, 144. 



As to experimental results it may be said that the work of van Calcar and 

 de Bruyn is in complete disagreement with the theory. The result reported 

 by the Earl of Berkeley and Dr. Burton is, however, of the predicted order 

 of magnitude. 



