250 



llete reaction to take place. Care must be taken during the last stages 

 (if tlae reaction, since tlie magnesite is used up and the HCl gas given off 

 lieing extre)nely soluble allows the potassium iodide solution to suck back 

 into the flask. This is prevented by heating. Heating at this iK)int not 

 only prevents the sucking back of the KI solution, but is necessary in 

 order to expel the last traces of chlorine from the flask. A second Nessler 

 tube should be inserted and the heating continued a few minutes, in order 

 to make sure that the reaction is complete. During the distillation tlie 

 Nessler tube is surrounded by a beaker of cold water, in order to keep 

 the temperature of the potassium iodide solution as low as possible so as to 

 l»revent the volatilization of the iodine. 



This method is both quick and accurate. A series of analysts on one 

 sample gave the following results: 31.84, 31.92, 31.92 per cent, lead per- 

 oxide. In all about fifty different samples were run by this method and 

 in every case it was easy to check the results to within 0.10 per cent. 



If a distilling flask with a ground glass stopper is not at hand, one 

 can be made in a few minutes. Select a glass stopper of the proper size 

 to fit the neck of the flask and fasten it in an horizontal position to the 

 end of a slowly rotating shaft or axle fit the flask over the rotating stop- 

 per and grind with fine emery dust moistened will a mixture of equal parts 

 of ether, turpentine, and alcohol. 

 Chemical Lahoratori/, 

 I II (lid II a Vnivcrftitij, 



