JOHN H. NORTHROP 631 



is proportional to the resistance per cm., it is possible to calculate the 

 drop in potential in the cell itself, provided the dimensions of the rest 

 of the apparatus are known. Expressed as an equation, 



U 



A 



Drop in potential per cm. in the cell = , X ^ot^^ drop in potential 



_ _i_ —4- LlL 

 Ac At At' 



Lc = length of cell in cm. 



A<. = area of cross-section of cell. 



Lt = length of tubing in cm. 



At = area of cross-section of tubing. 



The total potential is measured by a voltmeter connected to the Zn. electrodes. 



In experiments with salt concentrations of less than tenth normal the 

 resistance of the zinc sulfate is negligible. In any case the uncertain- 

 ties due to the change in dielectric constant with increasing salt con- 

 centration are probably greater than the error introduced by neglect- 

 ing the zinc sulfate. It can be seen from the above that failure to 

 allow for a widening of the system at some point would make little 

 or no difference in the potential gradient but that a narrow place in 

 the system even though very short would cause a very large error. 

 It is important, therefore, to be sure that the cell is not narrowed at 

 the ends by the cement or by failure to align the ends of the cell and 

 the side tubes. 



Method of Measuring the Velocity of Migration. — The cell is clamped 

 in position under the microscope, after filling the electrode tubes with 

 saturated zinc sulfate, and the electrodes connected to a source of 

 potential. The stop-cocks are turned so as to close the tubes con- 

 taining the zinc sulfate and the cell filled with the suspension, care 

 being taken to avoid air bubbles. The stop-cocks are then turned so 

 as to connect them with the zinc sulfate solution, the circuit closed 

 and the time required for a particle to cross a division of the microm- 

 eter eye-piece determined with a stop-watch. Owing to the migra- 

 tion of the water itself, it is necessary to obtain the average motion of 

 the particles in the cell as a whole. This may be done accurately by 

 determining the speed at various depths, say every 0.05 mm., plotting 

 the curve so obtained and determining the mean height from the arc a 



