^36 On deftnite PfeportiofiSi 



carbon, that is } per CQnt., or accurately '497 per cent, of 

 the quantity of the iron. The solution gave after filtration 

 •005 gr. of light irray silica. Five grammes of the same vvirc 

 dissolved in nitric acid, in a glass flask, then dried and ig- 

 nited in the flask, aflbrded 7*19 gr. of red oxide of iron, of 

 43"8 of oxvaen to 100 of iron. 



6.) Three grammes of the same wire were dissolved in 

 nitric acid, and precipitated by caustic ammonia. The pre- 

 cipitate, after ignition, weighed 4'305; so that it cou- 

 tamed 43'3 of oxvgen to 100 of iron. > 



The greatest of these results is that of No. 5, which gives 

 44-5 ot oxvgen to 100 parts of iron free from carbon. All 

 the rest vary but little from 44*25, so that the whole uncer- 

 tainty lies between 44| and 44|. I have assumed through- 

 out this essay 441 as the true proportion of oxygen in the 

 red oxide with 100 of iron. 



It is a remarkable inference from this expefiment, that 

 our common and most malleable iron still contains carbon, 

 which however does not exceed the proportion of | per 

 cent. In ill-manufactured bar iron, the quantity is cer- 

 tainly much greater, and hence may arise imperfections, 

 where no foreign substance can be discovered by the usual 

 modes of analysis. The presence of silica, or rather of its 

 base, in malleable iron, may also serve as a proof, that if 

 an ore of iron contains an oxide of any other metal, this 

 metal, which must alwavs be less combustible than silicium, 

 must also continue mixed with the cast iron, and even not 

 be completely separated in the manufacture of bar iron, but 

 always remain mixed with it in a small quantity. And the 

 same must be true of sulphur and phosphorus. Hence it 

 appears how necessary it is. in the examination of the faults 

 ot iron, to have at hand the ores and the fluxes employed 

 in its preparation ; since substances present in such minute 

 quantities cannot possibly be discovered, unless our atten- 

 tion has been particularly directed to them. 



B. Protoxide of Iron. 



It is generally believed that iron has only two degrees 

 of oxidation, that of the black and that of the red oxide; 

 and that the ildck, the martial aethiops, or finery cinder, 

 is that into whtch iron'is changtd durmg its solution in 

 acids. Si-vcral circumstances however appear to contradict 

 this opiui(m. 



It. for example, we add caustic ammonia to a recently 

 prtjMMd solution of iron in muriatic oi sulphuric acid, and 

 pie\tni the access ot aiinosphcric air, evei\ a considerable 



excess 



