ItUelligence and Miscellatteous ^4rticles. 69 



dently remains combitied in the iron, with it or other elements, ol* it 

 could not be eliminated as a hydrocarbon when the iron is treated 

 with acids; but the nature of these combinations of carbon would 

 be an ample subject for distinct investigation. In distinguishing 

 between the carbons existing in the specimens, I would only observe 

 that the 6-carbon is separated by acids, and may be collected on a 

 filter, whilst the a-carbon cannot be &o collected ; it escapes as a 

 hydrocarbon, having a powerful alliaceous odour, and may be par- 

 tially collected as a liquid. I have only further to notice the in- 

 variable presence of sodium or potassium, or both, in these irons. 

 Karsten states that they have not been found as constituents of iron 

 hitherto. As they are present in only a few of the ores, they are 

 probably, in part at least, derived from the smelting materials. 



Method of Analysis for the Irons. — The specimens were easily 

 broken into small pieces in a steel mortan In one portion of from 

 20 to 30 grs. the sulphur and phosphorus were determined. In a 

 second quantity, all the other constituents were determined except 

 the carbon. On being treated with HCl and warmed, the iron is 

 quickly acted on, and in a few hours dissolved, leaving black flakes 

 and particles floating in the liquid. These were collected on a filter 

 previously dried at 212° and weighed. After well washing, until 

 no trace of HCl remained, the filter was again dried and weighed. 

 The increase was carbon principally, with small quantities of sili- 

 cates of oxides of iron, lime, &c., and in the numbers VII. and VIII., iron 

 in an equivalent proportion to the carbon. In these latter, as also in 

 the numbers I. and III., the silica, iron, &c. of the substance sepa» 

 rated by the filter were determined by fusing it with nitrate of pot- 

 ash mixed with twice its weight of carbonate of soda; the iron, &c., 

 separated in the usual manner, gave the carbon by loss. This was 

 afterwards verified in two instances by a direct determination of the 

 carbon in a combustion-tube. The iron, &c., separated in the 

 above manner, was added to that obtained from the solution. The 

 carbon was designated b, and being deducted from the entire quan- 

 tity found in the iron by the method to be detailed, gave the quan- 

 tity of carbon designated a for the reason before named. The fil- 

 tered liquid and washings, evaporated to dryness and again treated 

 with acid and water, usually left a minute portion of silica, which 

 was separated, weighed, and added to the former quantity. A cur- 

 ;^ rent of sulphuretted hydrogen being passed through a small quantity 

 f^ of the solution, it in no case gave any other than a milk-white pre- 

 cipitate of sulphur. After being carefully freed from this and from 

 SH, it was returned to the main solution, NO^ added, and boiled 

 until all the iron was peroxidized, and ammonia added gradually 

 until the solution only faintly reddened litmus, and nearly all the 

 iron was precipitated. A little neutral benzoate of ammonia sepa- 

 ^ rated the last portions of the peroxide of iron. The precipitate, 

 after well washing, was dried, ignited, weighed and examined for 

 manganese by fusion with nitrate and carbonate of potash ; for chrome 

 and alumina by dissjolvtncMm kSlHV -tiBd-'brecipitatin^ Whh "<Jauiaitio 



1T9 daifiw Jfirii moil Ji deiygaiJaib oi ",iwvs odi moi) bsJfiiBqes " 



