217 



pieeos of zinc (■(HitaiiU'd in ;i iiJis double pipe! for solids. The iiipet was 

 fillt'd with lioilod distilled water to displace the air. Tlie sulphuric acid 

 was added throu.i^h a .ijlass tube which entered through the opening for the 

 introduction of solids into the pipet. The h.vdrogen gas was allowed to 

 escape completely from the apparatus several times before any was saved 

 lor analysis. This form of generator very eit'ectively protected the hydro- 

 gen gas from the diffusion of air. The results are given in the following 

 table : 



The" following table shows the results which were obtained in the 

 analysis of gas I'esidues : 



Residue Oxygen Volume after ry- To al (ajb.n 



'■"~^-^-— '"^ {'ontraf'tion. Diox;dp. 



S2.1 IS.S 



81.4 17.5 



81.34 17.77 



The result in experiment 1 was obtained with the ordin.irv combustion 

 pipet. This value was taken as the standard. Experiment 2 in the table 

 shows 17.5 per cent, of carbon dio.xide. which was of course obtained by the 

 absorption of the carbon dioxide which was in the buret. This value must 

 be corrected for the amotuit of carbon dioxide which remained in the 

 quartz tube. The total gas residue after conibustion (including the air 

 originally in the quartz; tube) was 42.4 plus 2.795 ( the volume of the quartz 

 tube which was unoccujiied b.v platinum i which was 45.195. So 93.8, 

 (42.4-1-45.195), was the per cent, of the gas which was measured in the 

 buret. 



All of the results given in the tables show that the quartz tube is as 

 accurate as the combustion pipet. It was found necessary tO' pass air 

 through the quartz tube to remove the carbon dioxide produced by one 

 e.xperiment if another experiment was to be made at once. If only total 

 contraction was desired the carbon dioxide did not nee<l to be removed. If 

 nitrogen was to. be determined the gas remaining in the tube from a previ- 

 ous experiment had to be removed by ]iassing air. 



Tlais apparatus gave an excellent method of determining the total 

 nitrogen in gas. The gas was mixed with an excess of oxygen whose nitro- 



15—33213 



