364 Kreider — Oxygen in Air and in Aqueous Solution. 



Nitric oxide was generated very satisfactorily according to 

 Professor Gooch's metliod — by the action of nitric acid upon 

 globules of copper in a Kipp generator. When the nitric acid 

 is diluted with an equal volume of water the evolution of the 

 gas is sufficiently rapid without the application of heat, but 

 contamination by the higher oxide is more likely. However, 

 since it is necessary, in order to be certain of purity, to pass the 

 gas through an acidified solution of potassium iodide before 

 applying it to the determination of oxygen, whatever higher 

 oxide may be present will be reduced. By passing the gas, as 

 it issued from the generator, through a set of Geisler bulbs 

 containing an acidified solution of potassium iodide and wash- 

 ing with potassium iodide solution, the perfectly purified gas 

 was obtained. Theoretically, only a small amount of the nitric 

 oxide is required for the transference of the oxygen to the 

 hydriodic acid, but when too little is taken the action is very 

 slow. On the other hand, too large an amount relieves the 

 vacuum to such an extent as to interfere with the introduction 

 of the air. A little device to measure the volume of gas taken 

 was therefore attached to the generator. It consisted of a 

 tube filled with water and roughly graduated for every five 

 cubic centimeters, so attached to the generator that the gas 

 would enter by displacement of the water, which would descend 

 to a lower bulb, and as the gas was withdrawn the water would 

 again take its place. Fifteen cubic centimeters of the gas was 

 found a convenient and satisfactory amount for the analysis. 



Carbon dioxide was generated in a Kipp generator, the 

 acid and marble of which had been previously boiled and 

 contained a little cuprous chloride. To remove a trace of 

 reducing matter which the gas was found to contain, it was 

 first passed through a solution of iodine and washed with 

 potassium iodide. 



For the titration a decinormal solution of arsenious oxide 

 (4'95 gms. to the liter) was employed : one cubic centimeter 

 being equal to 0-559846<="^' of oxygen at 0° and 760"^°^ when the 

 weight of a liter of oxygen at 0° and 760 "^°^ is taken as 1-42895 

 gms. When the volume of air taken is 100'^°'^ under standard 

 conditions of temperature and pressure, as obtained by Lunge's 

 device, the following table, calculated for the volume of oxygen 

 equivalent to the volume of arsenic solution, shows directly the 

 percentage of oxygen corresponding to the reading of the 

 burette. The correction necessary for the fraction of a tenth 

 of a cubic centimeter of the arsenic solution is obtained with 

 sufficient accuracy by simply multiplying by 0-006. 



