On definUe Proportions. S3S 



»nd not perfectly transparent, but dc{>ositcd no precipitate 

 alter standing a few hours. It was mixed with nitric acid 

 and boiled, in order con)pletely to oxidate the protoxide: 

 the caustic ammonia then afforded a precipitate, which 

 when washed, dried and igniled, weighed 5'74 gr. Coii?e- 

 x]uently 4gr. oi this iron had gained 1-74, iOO parts having 

 taken iip 43-3 of oxygen. There was no reason to attribute 

 any part of this addition to the solvent, or to the precipitant, 

 both being volatile ; and if they had not been perfectly washed 

 away, they must have volatilised aho some of the oxide, 

 Bucholz obtained from 100 parts of iron only 142 of the 

 red oxide; so that either some circumstance in his mode of 

 conducting the experiment must have occasioned a loss, 

 or his iron must have contained much more carbon than, 

 njine. 



If now we subtract the carbon contained in the iron, 

 there will remain 3-98 gr. of the pure metal, which afforded 

 5-74 of red oxide; and 5-74 : 3*98= 100 : 69-34, so that 

 the oxide of iron consists of 



Iron . . . 69-34 100-00 



Oxygen 30-60 4 }-23 



This subject was so important as to require that the ex- 

 periment should be repeated. In order to obtain iron of a 

 uniform quality, I had a large nail filed clean and divided 

 into several pieces. 



2.) A piece of this nail, which weighed 7-1 grammes, 

 was dissolved in diluted sulphuric acid, and afforded 117 

 decimal cubic inches of gas, which, when burnt in oxygen,^ 

 produced '235 gr. of carbonate of lime, containing -0344 of 

 carbon, or not quite \ percent, of the quantity of iron. 

 In the solution a black powder was cleposited, which 

 after drying weit;hed -006 gr. and was found to be silica 

 blackened by a little carbon. 



3.) Of the same nail five gr. were di?solred in pure nitric 

 acid in a glass flask, evaporated to dryness, and ignited in 

 the flask. "The oxide obtained weighed S-0023 gr. Con- 

 sequently IOO parts of the iron had taken up 43-25 parts of 

 oxvgcn. 



4.} Another portion, weighing 3*5 gr. was dissolved in 

 aqua rtgia, and precipitated by caustic ammonia. Tho 

 precipilcitc, after ignition, weighed 503 gr. ; so that 100 

 parts of iron had taken up 43-73 of oxygen. 



5.) I dis;«oKed .^-6 gr. of a thick polished iron wire in 

 muriatic acid. The gas obtained amounted to 93 decimal 

 cubic inches, and affcH-ded, after it had been burned in oxy- 

 gen, -BiJo ^r, ci" carbonate of lime, containing •027y gr. of 



carbon. 



