ANALYSIS OF NITRE. 253 



curate, but is most readily applied in practice, depends 

 upon the circumstance that a solution of pure nitre, at 

 the temperature at which it is saturated, is still capa- 

 ble of dissolving other salts, especially chloride of 

 sodium. 400 grms. of the powdered specimen are 

 shaken with 500 cub. cents, of a solution of pure 

 nitre; the salt is then filtered off, again washed with 

 250 cub. cents, of a saturated solution of nitre, dried 

 at 100, and weighed. The loss of weight expresses 

 the amount of the foreign salts. Since this process is 

 liable to error from many causes, and gives the amount 

 of pure nitre, on an average, 2 per cent, too high, 

 these 2 per cent, must not be neglected in calculating 

 the amount of impurity present. 



The following process is more accurate, which con- 

 sists in converting the nitre contained in any specimen 

 into carbonate of potassa, the amount of which is then 

 determined by means of the standard acid, as in testing 

 potashes. 



9.475 grms, of pure nitre furnish a quantity of car- 

 bonate potassa, which is capable of saturating 100 

 measures of the acid mentioned in the testing of pot- 

 ashes ; so that if this amount of impure nitre be em- 

 ployed, the number of measures of acid indicate at 

 once the percentage of pure nitrate of potassa in the 

 specimen. 



One-fourth of the above quantity (2.369 grms.) of 

 the crude nitre is weighed out, intimately mixed with 

 1 grm. of ignited lamp-black, or finely pulverized 

 graphite,* and with 12 grms. of ignited and finely-pow- 

 dered common salt, which serves to moderate the vio- 



* If common coal is used cyanide of potassium and cyanate of 

 potassa may be formed. Pure graphite may be prepared by mix- 

 ing Ceylon graphite with ^ of chlorate of potassa to which con- 

 centrated sulphuric acid is added, and then warmed until no acid 

 fumes are given off. The mass is then shaken with water, the 

 graphite washed and ignited. 



22 



