114 Chester and Cairns — Determination 



The crucible may be conveniently covered by two pieces of 

 platinum foil, lapping a little at the middle, and with the cor- 

 ners bent down over its edges to keep them in place. A hole 

 through one of them admits the end of the tube conveying car- 

 bon dioxide. The other covering piece, lapping over the first, 

 has one corner cut away a little to admit a small platinum 

 spatula, or stiff wire, which remains in the crucible during the 

 whole operation. 



A bent tube of hard glass, enlarged a little near the end, which 

 pas%es through the hole to prevent it from slipping in too far, 

 conveys the carbon dioxide from a generator, or more conven- 

 iently, from a gasometer, to the crucible. Between the gas- 

 ometer and the crucible a washing bottle charged with concen- 

 trated sulphuric acid should be placed. 



The pulverized mineral is put into the crucible on the water 

 bath, which should be kept boiling during the whole operation, 

 the carbonic acid turned on and allowed to drive the air out. 

 The flow may then be reduced to about two bubbles per sec- 

 ond, which is quite sufficient to keep the crucible filled with 

 the gas. A few cubic cent, of concentrated sulphuric acid are 

 then stirred up with the powder, and finally a little ammonium 

 fluoride is added, causing a violent frothing of the materials. 

 After this action has ceased more of the reagents are added in 

 several portions until such an addition causes no apparent 

 change. The operation requires about 25 mg. of the fluoride 

 for every 10 mg. of silica in the sample; that is, for 5 gm. 

 of a silicate containing about 50 per cent of silica 1*25 gm. of 

 fluoride will be needed, added in three or four portions, at in- 

 tervals of five to ten minutes, sufficient sulphuric acid being 

 added with it to keep the mass fluid. It is desirable to stir it 

 up frequently, particularly during the first part of the process, 

 and for this reason, as well as for convenience in watching the 

 operation, the cover is made in two pieces, so that one can be 

 readily removed and replaced. 



Decomposition being complete, the contents of the crucible 

 may be poured into a large beaker containing cold water, the 

 crucible and spatula washed into it, the solution properly diluted 

 and titrated as usual. In cases where the mineral contains 

 much ferric silicate, a basic salt, insoluble in cold water, will 

 be formed. If desired, the beaker may be filled with carbon 

 dioxide, and the contents heated until a perfectly clear solution 

 is obtained, but this is not necessary, for all the ferrous sulphate 

 dissolves in the cold water, and the milkiness produced by the 

 other does not interfere in the least with the delicacy of the 

 color reaction, but rather aids it. 



Several experiments were made to test the accuracy of this 

 method with results given below. Two sources of inaccuracy 



