DETECTION OF IMPURITIES. 273 



The presence of a small fraction of one per cent, of sodium chloride 

 in many official chemicals cannot be looked upon as objectionable, 

 while the same amount of arsenic would render the preparation unfit 

 for medicinal use. 



The methods used by the Pharmacopoeia to determine the quality 

 of a chemical preparation may be divided into four classes, as follows : 

 1. Tests as to identity ; 2. Qualitative tests for impurities ; 3. Quan- 

 titative tests for the limit of impurities ; 4. Quantitative determina- 

 tion of the chief constituent. 



Tests as to identity. These tests are partly of a physical, partly 

 of a chemical character. They include, in the physical part, the 

 examination of the appearance, color, crystalline structure, specific 

 gravity, fusing-point, boiling-point, etc. 



The chemical tests given are sufficiently characteristic to leave no 

 doubt as to the true nature or identity of the substance. In order to 

 accomplish this object it is not necessary to apply all the analytical 

 reagents characteristic of the substance or its component parts, but 

 the U. S. P. selects from the often large number of known tests one, 

 or possibly a few, which answer best in the special case. 



For instance, while we have a number of tests, both for potassium 

 and iodine, the U. S. P., in the article on potassium iodide, gives but 

 one reaction for each of these elements. Yet these tests have been 

 selected with sufficient judgment to admit of no doubt regarding the 

 nature of the substance. 



Qualitative tests for impurities. These tests are in many cases 

 described minutely, i. e., the quantity to be taken of both the sub- 

 stance to be examined and the reagent to be added is stated. More- 

 over the amount of solvent (water, acid, etc.) to be used is mentioned, 

 and other details are given. The object of this exactness in describ- 

 ing the tests is not only to render the work easy for one not fully 

 familiar with analytical methods, but also, in some cases, to fix a 

 limit for the admissible quantity of an impurity. A certain reagent 

 may, in a concentrated solution, indicate the presence of a trace of 

 an impurity, while in a more dilute solution this reagent will fail to 

 detect it. The selection of the reagents used in certain tests is also 

 made with the view of establishing a sufficient purity for pharmaco- 

 poeial purposes of the article examined without demanding an absolute 

 purity. 



A few instances may help to illustrate these remarks : Potassium 



18 



