DETERMINATION OF ALKALIES IN SILICATES. 401 



of furnace and crucible is found to be convenient for other 

 operations. 



Although the details here given are long, the time occupied 

 in the analysis is short, and the necessary precautions are of a 

 simple character, so much so that results obtained by students 

 in beginning chemical analysis have been found by me far 

 more reliable and less variable on the alkalies of the silicates 

 than on any of the other constituents. Good alkali determina- 

 tions can be made in three hours or less from the commencement 

 of the operation, hastening the evaporation by more direct appli- 

 cation of the heat, which of course requires more close watching. 



It is a common practice of mine, when a silicate presents 

 itself of which there are no physical means of ascertaining its 

 nature, to make at once an alkali determination, which not 

 unfrequently indicates immediately what it is, if it be a known 

 silicate containing an alkali; and if an unknown compound, the 

 analysis made is one step in the examination. 



