dependent on the Motion of the Conductor. 97 



the liquid, becomes therefore 



-fie* ....... (9) 



This deflection is consequently inversely proportional to the 

 resistance of the liquid ; while the deflection occasioned by the 

 change in the resistance is, according to equation (1), nearly 

 independent of the same resistance. If, then, liquids of suffi- 

 ciently great resistance be selected for the experiments, we can 

 make the variations in the polarization harmless without thereby 

 lessening the deflections produced by the alteration of the re- 

 sistance. Moreover it follows from equation (1) that the de- 

 flections in question are proportional to the cross section of the 

 glass tube and to the velocity of the liquid, but independent of 

 the intensity of the galvanic current employed. 



Of the great number of observations conducted in the way 

 above described, it may be sufficient for the purpose before us 

 to communicate only the following. 



III. 



For the experiments a cylindrical vessel of thick sheet cop- 

 per was used, tinned on the inside, and capable of holding 

 about 25 litres. The upper end was hemispherical and pro- 

 vided with three apertures. In one of them a cock was fixed, 

 which was connected with an air-condensing engine by means of 

 a leaden pipe. In order to measure the condensation of the air 

 produced by the engine, a manometer could be screwed on one 

 of the other apertures, which was also made use of for the pur- 

 pose of filling a part of the vessel with liquid. In the third 

 aperture a copper pipe with a cock was fixed air-tight. That 

 part of this pipe which was inside descended perpendicularly 

 nearly to the bottom of the vessel ; the part outside was bent 

 the form of a semicircle, so that its extremity descended perpen- 

 dicularly. To this end of the copper tube the above-described 

 glass tube was screwed. The liquid flowing through the glass 

 tube was received in a vessel of glass or wood standing on the 

 floor of the room. Before each experiment, air was forced 

 into the vessel, above the liquid, up to a pressure of two or 

 three atmospheres above the pressure of the external air. No 

 measures were adopted to keep the pressure constant and in- 

 variable ; and on this account the velocity of the liquid grew less 

 during the outflow. This inconvenience, however, was of little 

 consequence, because in this investigation the question was 

 rather to obtain an experimental proof of the dependence of the 

 resistance on the motion of the liquid than an accurate deter- 

 mination of the laws of that dependence. 



