356 



PROCEEDINGS OF THE AMERICAN ACADEMY. 



been worked upon are gold and bismuth. In preparing these for the 

 measurement of the Hall effect he used the form of plate employed 

 extensively by Righi ; see P, Figure 1. 



Thus the current entering the plate at A was divided into the two 

 partial currents at i? and B. On the introduction of the magnetic 

 field one partial current was increased at the expense of the other above 

 its normal value, but correspondingly decreased on the reversal of the 

 field. 



At c and c', d and d' , connection 

 was made with the voltameters 

 which were placed in each branch. 

 The voltameters were of the fol- 

 lowing form. See Figure 2. 



The jar ^^ contained a saturated 

 solution of copper sulphate, the jar 

 CD a solution of ten per cent by 

 weight of sodium sulphate. The 

 ends of the curved tube were cov- 

 ered with pieces of animal mem- 

 brane. It was filled with a solution 

 of sodium sulphate. The current 

 was made to pass from the copper 

 sulphate solution to the sodium 

 sulphate solution. The amount of 

 sodium hydroxide formed at the 

 cathode was measured by a method 

 of volumetric analysis suggested by 

 Cardani,* for use in measuring the 

 quantity of electricity in the dis- 

 charge of a condenser. 



Cardani added measured quan- 

 tities of a ^Jo normal oxalic acid 

 solution to a nine per cent sodium chloride solution which had been 

 around the cathode while the current was passing. Using phenolphta- 

 lein as indicator, he endeavored to obtain by titration the same degree 

 of redness as was shown by the original weakly alkaline sodium chlo- 

 ride solution before the passage of the current. In Moretto's work the 

 strengths of the two partial currents were measured, first without the 



* Nuovo Cimento, Serie 4, 7, 105 (1898); also Zeitschrift fiir physikalische 

 Cheniie, 27, 378 (1898). 



Figure 2. 



