AMERICAN INSTITUTE. 601 



to escape, the materials are properly mixed, and become, in turn, 

 equally exposed to the heat and to the draft. 



3. The draft is unconfined, and moves freely and rapidly for 

 carrying ofif the impurities. 



4. The atmospheric air is deprived, for the most part, of its 

 oxygen by the fuel of the fire-place, and, therefore, while passing 

 rapidly through the ores, it does not oxydize the metal, and does 

 not consume the carbon, which is consequently allow^ed freely to 

 extract the oxygen from the oies. By the gradual heating and 

 freedom of draft and frequent agitation, an opportunity is afforded 

 for the free escape of impurities in their natural order, beginning 

 with the more volatile, and ending with the more fixed. Such 

 escape of gaseous products is more difficult while a mass of solid 

 materials, from which they are generated, remains at rest. 



5. The agitation may be carried on at different temperatures, 

 so that the objects which it cannot effect at one degree of heat it 

 w411 at another. This is the purpose of the three several cham- 

 bers, of which the upper is the heating and vaporizing, the middle 

 the mixing, and the lower the reducing and finishing chamber. 



It is alleged that the ores can be reduced to metals of more 

 than ordinary purity by the above-mentioned means. The ores 

 of iron may be reduced to wrought or malleable iron without first 

 carbonizing the iron. They maybe reduced also to a carbonized 

 state, either as steel, or as cast or pig iron ; this may be done by 

 having less agitation, and adding an excess of carbon. 



The necessary materials for the reduction of the ores may be 

 introduced at different temperatures, and at different stages of 

 reduction, according as their presence maybe needed. For ex- 

 ample, when lime is required for separating silica from iron ore, 

 such lime need not be introduced at the beginning of the process, 

 wdien the temperature is low, for at such temperature it cannot 

 act upon the silica, and its presence w^ould certainly interfere 

 with the free expulsion of other impurities. It may, therefore, 

 be introduced partly in the middle, and partly in the lower 

 chamber, as needed. 



The carbonic acid gas evolved from the limestone or shells in- 

 troduced in the lower chamber, tends to protect the carbon and 

 ores and impurities from the residuum of free oxygen left in the 

 draft. 



It is claimed that this process yields a greater percentage of 

 metal from any given amount of ote than is obtained b}' other 

 furnaces heretofore used. The ores, and the necessary materials 

 for their reduction are, through the whole process, completely un- 



