360 Dr. J. W. Capstick. 



P. E. 



12 370 



9i 372 



7 368 



4 360 



3J 364 



3 374 



2i 374 



2 374 



li 373 



1 373 Mean, 370 



The mean of the whole is 370 volts. 



Nitrogen. Warburg's determination of the kathode fall in this gas 

 was made on atmospheric nitrogen containing argon, and hence it 

 was necessary to repeat the observations on chemically prepared 

 nitrogen. 



In the first experiments the gas was prepared by heating ammonium 

 bichromate. If the bichromate is mixed with fine sand the decom- 

 position is easily controlled, but the gas is impure, and must be 

 passed over ignited copper and copper oxide, which involves the 

 use of hard glass and india-rubber connections. The joints were very 

 carefully made with thick-walled soft rubber tube, yet the observed 

 kathode fall, 355 volts, was the same as in air, and nearly the same as 

 in oxygen, and very much higher than what later experiments showed 

 was the true value for nitrogen. The result is interesting, as showing 

 the great effect of a small quantity of oxygen in the nitrogen, and 

 emphasises the necessity for scrupulous care in removing traces of 

 impurity. 



The final experiments were made on gas prepared by the decom- 

 position of ammonium nitrite. A solution of pure ammonium 

 chloride was contained in a flask from the neck of which a tube a 

 yard long passed downwards into a mercury reservoir. When the 

 flask was exhausted along with the rest of the apparatus, potassium 

 nitrite solution could be sucked in as required by pouring a little on 

 the top of the mercury, and lowering the reservoir. At low pressures 

 the mixture of ammonium chloride and potassium nitrite does not 

 require heating, as the nitrogen comes off regularly and very slowly 

 at the temperature of the room. 



The gas was passed over solid potash and phosphoric anhydrid< 

 and a piece of sodium was placed in the discharge tube. 



In some of the experiments bichromate of potash was also added 

 to oxidise to nitric acid any oxides of nitrogen that might be 

 formed. The addition made the evolution of gas inconveniently 

 rapid, and the flask had to be kept cool by a water bath. The 



