SPIRIT OF NITROUS ETH^R. ' 153 



Referring to table 26 it will be seen that the result obtained 

 by the new United States Pharmacopoeia method is in every 

 case the lowest, and that the result obtained by the modified 

 method is in each case the highest, when compared with the re- 

 sults by the other methods on the same sample. That this 

 difference represents an actual gain in accuracy for the modified 

 method seems apparent when the work and results are carefully 

 analyzed. 



The manner in which the determination is made by the 

 modified method when absolute accuracy is desired is as fol- 

 lows : The nitrometer shown in the illustration, Fig. 268, is 

 filled to the top of the stem of the thistle tube with a saturated 

 salt solution, which is drawn down to the bottom of the ungrad- 

 uated arm by means of the outlet tube at the bottom after the 

 stop-cock at the top has been closed ; thus leaving the graduated 

 arm full but under reduced pressure. The bottle containing the 

 sample is weighed and then five cubic centimeters are with- 

 drawn in a pipette and the bottle again weighed. This five c. c. 

 portion is placed in the thistle tube, drawn down, and washed 

 in with a few drops of alcohol. Then in succession ten cubic 

 centimeters of a ten per cent solution of potassium iodide and 

 ten cubic centimeters of a five per cent solution of sulphuric 

 acid are introduced in a like manner, taking care that no air 

 is accidentally drawn in during the process. In order to obtain 

 all of the five c. c. portion of the sweet spirit of nitre which 

 was drawn from the bottle the potassium iodide solution is 

 allowed to run through the pipette and wash out into the 

 nitrometer every trace of the sample. Nitrogen dioxide gas, 

 NO, is immediately formed, and, by means of the outlet tube 

 at the bottom, the salt solution is drawn down so that the level 

 of the liquid in the ungraduated branch is kept below that in 

 the graduated. The reaction is rapid for the first two or three 

 minutes and then proceer\s slowly, for an indefinite period. The 

 process should be allowed to continue for one hour with fre- 

 quent shaking and the volume of the gas is then read with the 

 liquid in the two tubes at the same level. This volume multi- 

 plied by 0.0030673, the product multiplied by 100, and then 

 divided by the weight of the sample taken, gives the per cent of 

 ethyl nitrite present after the result is corrected for tempera- 

 ture and pressure. 



