596 Viscosity of Hydrogen, Nitrogen, and Oxygen. 



It should be pointed out here that even this analysis was 

 only a qualitative one. Since the degree of purity required 

 was much higher than that which could be reached by this 

 method of analysis, and since no accurate spectral analysis 

 of oxygen is possible, it had to be assumed that the gas was 

 pure ; and with the method and procedure followed in the 

 production and purification such assumption was not by any 

 means unjustifiable. 



Table III. contains the results of the individual obser- 

 vations and their mean. As in the case of hydrogen and 

 nitrogen, the observations were made with three changes of 

 the content of the chamber. 



Summary. 

 .1. The viscosities of hydrogen, nitrogen, and oxygen were 

 determined by means of the constant deflexion apparatus 

 which was designed and developed by Professor Millikan. 



2. These determinations were made at a temperature 

 of 23 u '00 C. and under a pressure of 76 cm. mercury. 



3. The results obtained for these gases were : — 



Hydrogen. . . . v x 10 r = 88216 

 Oxygen .... „ = 2042'35 



Nitrogen .... = 1764-80 



each with a probable error of 0*15 of one per cent. 



In conclusion the writer wishes to say that it was under 

 Professor Millikan's direction that these determinations 

 were made, and that it was the University of Chicago that 

 defrayed all the expenses involved in the experiment. 



Addendum. — Since there is in existence an enormous 

 amount of literature concerning the subject of viscosities, it 

 was thouoht advisable to ^ive as references only those works 

 that are closely connected with this particular method. 



The several methods for the production and purification of 

 gases were found in Grmelin, Kraut und Friedheim, Uandbuch 

 der Anorganischen Chemie. The. employment of metallic 

 calcium for the absorption of nitrogen and oxide of copper for 

 the removal of hydrogen was found in Dennis's ' Gas Analysis/ 

 The particular form of arrangement and order of procedure 

 were results of what is known as "learning by trial and error." 



Ryerson Physical Laboratory, 

 University of Chicago, 



Chicago, 111., U.S.A. 



July 3rd, 1919. 



