MAGNETITE. 45 



and the liquid heated to ebullition. It is 'retained at 

 this temperature until the dark-brown color originally 

 observed has changed to a pale yellowish-green. The 

 whole of the iron is now contained in the solution as 

 protochloride, in consequence of the formation of sub- 

 chloride of copper. The orifice of the little tube is 

 closed air-tight, and the solution allowed to cool some- 

 what. The flask is then filled with hot water, the liquid 

 poured off' from the undissolved copper, which is to be 

 washed, first with dilute hydrochloric acid, then re- 

 peatedly with water, dried, and weighed. The atomic 

 weight of copper is to that of iron as the quantity of 

 copper dissolved is to that of the iron sought. 



In order to determine directly the amount of pro- 

 toxide and sesquioxide of iron present in. a substance, 

 it must be dissolved in hydrochloric acid. The follow- 

 ing is the method adopted: The compound is dissolved 

 in an excess of concentrated hydrochloric acid, in a 

 flask filled with carbonic acid, and afterwards closed; 

 the flask is then nearly filled up with water, previously 

 boiled, and a weighed strip of copper introduced; the 

 closed flask is placed in water, which must be gradually 

 heated to boiling, the subsequent process being con- 

 ducted and the result calculated as directed above. 



Or the weighed substance is placed in a flask closed 

 with a cork, and furnished with tubes for ingress and 

 egress, and with a funnel-tube passing to the bottom 

 of the flask, which is to be filled with carbonic acid. 

 Hydrochloric acid is then added through the funnel- 

 tube, and the solution assisted by heat, whilst carbonic 

 acid is allowed to stream through the apparatus. The 

 solution is afterwards diluted, through the funnel-tube, 

 with boiled water, and a milky mixture of carbonate 

 of baryta with water gradually added; this precipitates 

 the whole of the susquioxide of iron, while the protox- 

 ide remains in solution. When the supernatant liquid 

 has become clear, it is decanted through the egress- 



