110 PROCEEDINGS OF THE AMERICAN ACADEMY. 



pending upon the inertia of this matter, are to be expected. The 

 nature of these phenomena was completely discussed by Maxwell ; 

 but they have not yet been detected in metallic conductors.^ In elec- 

 trolytic conductors, however, the experiments of Hittorf have shown 

 that there is an actual transfer of matter through the solution, accom- 

 panying the passage of a current. As an example of the electrical 

 effects which may accompany such a motion of the carriers of elec- 

 tricity or ions, let us consider a solution of silver nitrate placed in 

 a vertical tube with silver electrodes at the top and bottom. The 

 passage of a current of electricity through this solution from the lower 

 to the upper electrode will be accompanied by a raising of the silver 

 ions in the solution and an approximately equal lowering of the con- 

 siderably lighter nitrate ions. The net etfect produced by the current 

 will be the lifting of a certain weight of material from the lower elec- 

 trode to the upper ; and the work thus done against gravity must cor- 

 respond to a definite electromotive force which will oppose the passage 

 of the current in this direction. 



Effects of this kind were first predicted by Colley,^ and have also 

 been considered by various other investigators.^ CoUey himself was 



^ Maxwell, Treatise on Elec. & Mag., 3rd edition, Vol. II, 211 et seq.; 

 Lodge, Modern Views of Elec. & Mag., 3rd edition, 89; Nichols, Phys. Zeitschr. 

 7, 640 (1906). 



^ Colley, Journal der St. Petersburg chem. und ph5's. Gesellschaften 7, 333, 

 1875; Pogg. Ann. 157, 370 (1876); Ibid. 157, 624 (1876); Wied. Beibl. 5, 

 457 (1881); Wied. Ann., 17, 55 (1882). 



3 Lodge, Phil. Mag. 2, 367 (1876); Hertz, Wied. Ann. 14, 590 (1881); 

 Des Coudrcs, Wied. Ann. 49, 284 (1893); Ibid. 57, 232 (1896). The electro- 

 motive force which is produced in salt sohitions by the action of gravity must 

 be carefully distinguished from the actual changes in the concentration of 

 the solute which gravity will |)roduce. The difference in concentration be- 

 tween the solution in the upper and lower ends of a vertical tube, or the 

 central and jieriphcral portions of a rotating tube can be calculated from 

 simple therinodj-namic considerations. It is to be exj)ected, however, that 

 this difference in concentration will be very slowly established (see Des Coudres, 

 I. c. below). The difference in potential between the two ends of the solu- 

 tion is an immediate phenomenon which occurs as soon as the tube is set up. 

 It is evident that when the final change in the concentration of a salt solu- 

 tion has completed itself there will no longer be any potential difference be- 

 tween the upper and lower portion of the solution. A complete bibli()grai>hy of 

 the theoretical and experimental work which has been done in this field follows: 

 Gay Lussac, Ann. chim. phys. 11, 306 (1819); Gouy et Chaperon, Ann. ciiim. 

 phys. 12, 384 (1887); Nernst, Zeitschr. f. phys. Chem. 2, 637 (1888); Duhem, 

 Journ. de phys. 7, 391 (1888); T. v. Turin, Journ. der russ. Geschiclito, 24, 

 90 (1892), Wied. Beibl. 17, 16 (1893); Des Coudres, Wied. Ann. 49, 284 

 (1893); Ibid. Wied. Ann. 55, 213 (1895); Van der Waals, Zeitschr. f. phys- 



