MEMOIR ON EMERY. 19 



once speak of the method which gave me very easily the most 

 accurate results. 



It is by means of the bisulphate of soda that all my analy- 

 ses of emery, of corundum, and of several aluminates were 

 made. I believe that I am the first who has shown the great 

 advantage of using this double salt in the decomposition of 

 certain substances insoluble in the acids; and very probably it 

 will replace in most cases the use of the bisulphate of potash 

 in analytical chemistry. At present all the advantages that 

 may arise from the substitution of the soda for the potash salt 

 can not be mentioned; all that I will say is that the former, in 

 giving a decomposition at least as complete as the latter, fur- 

 nishes a melted mass quite soluble in water, and in the future 

 operations of the analyses there is no embarrassment from a 

 deposit of alum. 



The bisulphate of soda was prepared by adding an excess 

 of pure sulphuric acid to the pure carbonate or neutral sul- 

 phate of soda, and heating it in a capsule until all the water 

 had been expelled and sufficient of the acid to allow of the 

 mass becoming solid on cooling. That obtained in commerce 

 is not sufficiently pure. 



The pulverized emery is placed in a large platinum crucible 

 with six or eight times its weight of bisulphate of soda, and 

 the mixture is heated over a lamp in the same manner and with 

 the same precautions as are employed when using the bisul- 

 phate of potash. From fifteen to thirty minutes suffice for the 

 operation. The mass is allowed to cool, and water with a few 

 drops of sulphuric acid are added to it, and the whole heated, 

 when it soon dissolves, with the exception of a little silica that 

 renders the solution milky, and a small quantity of undecom- 

 posed mineral that is readily detected by rubbing a glass rod 

 against the bottom of the capsule. The liquid is now filtered, 

 and the filter is washed once with a little water; then with its 

 contents it is placed in a platinum crucible, burnt completely, 

 and the residue is heated with a little bisulphate of soda, which 

 completes the decomposition ; and when treated with water 

 and a drop or two of sulphuric acid all except the silica is dis- 

 solved. The liquid which passes the filter in this case is added 

 to the first and the analysis continued. The silica obtained is 

 diminished by the quantity taken up from the mortar in order 



