258 CLARK: REDUCTION POTENTIALS 



the surface of the stopper; the calomel electrode arm drawn to a 

 fine gooseneck tip; and the electrode. The electrode was a 

 platinum wire wound as a spiral and plated with gold. The 

 electrode itself was kept immersed in the solution. It was fused 

 into a glass tube carrying a mercury contact. Since it was not the 

 intention to make a careful study of liquid junction potentials 

 these were reduced so far as is practicable by the customary use 

 of saturated potassium chlorid. This was accomplished by employ- 

 ing directly the "saturated" calomel electrode. Its potential 

 was frequently checked against that of four very constant and 

 carefully made tenth-normal calomel electrodes whose average 

 potential in terms of the normal hydrogen electrode^ was con- 

 sidered to be +0.3370. Both calomel electrode and titration 

 vessel were immersed in an oil bath maintained at 30° C. The 

 potentiometer equipment was that used in the writer's previous 

 work with the hydrogen electrode. Its principal features were a 

 Leeds and Northrup type K potentiometer and the same com- 

 pany's type R galvanometer with a megohm sensitivity of 1973. 



The reducing agent, as mentioned before, was titanium in the 

 presence of citrate and an excess of buffer. Since these solutions 

 are unstable in the presence of oxygen they were prepared by 

 adding the titanium to the previously prepared buffer mixture, 

 quickly bubbling oxygen-free nitrogen through it, and flushing 

 the whole system, burette and reservoir, with nitrogen. 



The titrating vessel was so arranged that it could be thoroughly 

 shaken after each addition of reducing agent. When this was 

 done electrode equilibrium was reached within a minute or two 

 except in certain of the more extreme conditions. 



In the following tables there are given in the first column the 

 number of cubic centimeters of reducing agent added. By plotting 

 these against the potential it will be found that at certain ^h 

 values there is not the sharp-end point that occurs at other 

 acidities. Such relationships will place upon a more quantitative 

 basis the empirical findings of Knecht and Hibbert, who dis- 

 covered that a sharp-end point in the titration of many dyes such 



* See Clark and Lubs. Journ. BioL Chem. 25: 479. 



