290 SILICATES. 



each part of the surface of the matter becomes incan- 

 descent, allowed to cool, again weighed, and a difference 

 is always found, provided the carbonate contains hygro- 

 scopic water. 



The material being thus prepared, it is heated for 

 fifteen or twenty minutes over the smaller lamp (gas 

 lamp fed by bellows) in such a manner that the car- 

 bonate acts upon the silicate without fusing; after 

 having thus expelled the carbonic acid, the substance 

 is placed over the large lamp, and it is necessary that 

 the glass produced should be well fused, homogeneous, 

 and if it is colored, transparent; all the peculiarities 

 should be observed, and the weight thus produced 

 should be determined. 



The glass should then be detached from the crucible 

 with the greatest possible care, in such a manner as 

 not to lose any of it, placed in an agate mortar, covered 

 with sheepskin, and ground with care, but not too fine. 

 The pulverized glass is then placed in a weighed pla- 

 tinum crucible, heated to 200 or 800 ; and the glass 

 to be analyzed weighed. 



The glassy material moistened with water is treated 

 with nitric acid, being stirred constantly with a glass 

 rod, so as to prevent the mixture forming a compact 

 mass at the bottom of the crucible. When all that is 

 found upon the glass rod is detached, and it is heated 

 over the lamp to be sure that nothing remains, the 

 crucible is placed upon the sand-bath and heated to 

 such a temperature that no more nitric acid is given 

 off and nitrous vapors begin to form. 



If the material contains any iron or manganese, it is 

 necessary to wait until the color becomes uniformly 

 red or black, and then there should be added enough 

 of a concentrated solution of nitrate of ammonia to 

 moisten the entire mass, which is heated over the 

 sand-bath, covering the crucible with a funnel ; after a 

 moment it is uncovered and odor observed. If the 



