TECHNIQUE OF ELECTROCHEMICAL MEASUREMENTS 465 



arising from this source, adopted the plan of heating the electrode, 

 after platinization, to a dull red, thus clumping the platinum black 

 and reducing the total surface of the electrode. 



I formerly tacitly assumed this phenomenon to be connected in 

 some way with the passage of the alternating current through the 

 solution. In my own experiments, however, I speedily found that 

 in neutral or faintly acid solutions of potassium caseinate marked 

 precipitation of casein occurs at the gas electrode, without the 

 passage of any current, usually after but sometimes even before 

 the passage of the hydrogen. At first I was inclined to attribute 

 this to gases collected, in the interval between experiments, in the 

 tubes conveying the hydrogen, to impurities in the hydrogen, etc., 

 but after very careful exclusion of all of these possibilities the 

 precipitation still took place, and it occurred to me that it might 

 be due to the hydrogen ions dissolved in the platinum itself. 

 When hydrogen is passed through a platinized platinum electrode, 

 the potential measured across the chain represents, at first, a 

 higher acidity of this electrode than it does later on, when the 

 electrode has come into equilibrium with the solution; in other 

 words, the hydrogen ions dissolved in the platinum have not yet 

 come into equilibrium with those in the solution; an excess of 

 hydrogen ions is still present in the platinum. It therefore 

 appeared possible that this initial acidity of the platinum itself 

 might be responsible for the precipitation of casein at its surface, 

 and this idea found confirmation in the fact that on prolonged 

 passage of hydrogen the film of protein deposited on the electrode 

 slowly redissolved. It occurred to me that it might be possible to 

 avoid this precipitation altogether by bringing the electrode nearly 

 to equilibrium with a low concentration of hydrogen ions before 

 introducing it into the solution at all; accordingly, before making 

 a gas-chain determination with a caseinate solution of very low 

 hydroxyl concentration (neutral or acid to litmus) the electrode 

 which was to be dipped into the protein solution was immersed in 

 distilled water and gas was passed through it for an hour or more 

 it was then immediately washed in the protein solution and used. 

 The device was found to work excellently, and in all of the experi- 

 ments described herein precipitation at the electrodes was avoided 

 entirely, at least so far as the eye could perceive. A similar device 

 was found effective in avoiding precipitation at the electrodes of 

 the conductivity-vessel. The vessel was simply filled with dis- 



