356 DR T. THOMSON on Asbestus, Chlorite, and Talc. 



The ratio between the lime and magnesia is the same as in 

 amphibole ; but there is a deficiency of silica amounting to 1 .8 

 atoms. But I believe that deficiency of silica to be only appa- 

 rent, and to be owing to the excess of 6 per cent, in the weight 

 of the constituents. Such an excess is very apt to take place in 

 the analysis of magnesian minerals, and is chiefly owing to the 

 formation of certain double magnesian salts, unless great care be 

 taken in precipitating the magnesia. I generally precipitate 

 the magnesia by carbonate of soda, and, after boiling the mixture 

 for some time in a flask, to drive off the excess of carbonic acid, 

 evaporate the whole to dry ness in a porcelain dish. The mag- 

 nesia remains when the residue is washed with water. This 

 magnesia is edulcorated, dried, heated to redness, and weighed. 

 Now, I generally begin my analysis of magnesian minerals, by 

 adding to the muriatic solution (formed after fusing the portion 

 of pounded mineral, subjected to analysis with carbonate of soda, 

 and getting rid of the silica), a quantity of bicarbonate of potash, 

 which throws down the alumina and oxide of iron, but retains 

 in solution the Lime and magnesia. This method was first prac- 

 tised by VAUQUELIN, during his analysis of the chrysolite. After 

 numerous comparative trials I adopted it as the most effectual 

 method of freeing the magnesia from alumina, and have accus- 

 tomed my practical students to employ it in their analyses of 

 magnesian minerals. The only objection to it is, that unless 

 care be taken, a double carbonate of potash and magnesia is apt 

 to be obtained, instead of pure carbonate of magnesia, as was 

 first pointed out by BERZELIUS. If we allow the formation 

 of a little of this double salt in the preceding analysis, then the 

 constituents of amianthus will agree exactly with those of some 

 varieties of amphibole. We may represent it as consisting of, 



