1835.] Improvements in Science. 115 



the chloride into muriatic acid. This method, however, is 

 not correct, because the whole of the chloride of magnesium 

 is not decomposed, and, by consequence, a portion of it is 

 obtained along with the soluble salt, and spirits act in the 

 same way as water. But when chloride of magnesium is 

 heated in a small platinum crucible, over a lamp with a 

 double stream of air, and then a bit of carbonate of ammonia 

 is placed in it, and the heating repeated several times, with 

 the precaution of moistening the salt with a drop of water, 

 the magnesia will be completely decomposed. If the dried 

 residue be dissolved, and tested with nitrate of silver, a 

 mere trace of the chlorine will only be detected. 



Rose endeavoured to separate in this way chloride of 

 lithium from chloride of magnesium, but of the original 

 quantity of chloride of lithium, he only obtained 93 per 

 cent. The remaining magnesia afforded an opalescence 

 with nitrate of silver. The cause of the difference is, that 

 part of the lithia, from the presence of carbonate of am- 

 monia, is converted into carbonate of lithia, which, on 

 account of its solubility, cannot be readily separated from 

 the magnesia. 



Chlorides of sodium and potassium may be heated 

 frequently with carbonate of ammonia, without changing 

 their weight. Chloride of calcium, on the other hand, 

 undergoes a similar change with the chloride of lithium. 



By access of the air, the chlorides of potassium and 

 sodium are changed, and therefore, the preceding state- 

 ment must be understood as referring to their behaviour in 

 a covered crucible. 



Rose has made some experiments upon the volatility of 

 chlorides of sodium, potassium, and lithium, which are 

 very important. 



1*0255 grms. of chloride of potassium lost by a red heat 

 in a small platinum crucible 6 lines in depth during the first 



quarter of an hour - 0*0845 



During the second quarter - - - - 0*0890 



1*016 chloride of sodium in the same vessel during the 



first quarter 0*038 



During the second quarter - - - - 0*039 



A mixture of 0*932 grms. chloride of potassium, and 

 1*175 chloride of sodium, lost in the same circumstances 



i2 



