222 ANNUAL OF SCIENTIFIC DISCOVERY. 



reactions, is obtained through the use of the voltaic battery to heat the 

 iron whilst it is exposed to the action of the atmosphere of certain gaseous 

 or volatile matters. 



Let the malleable iron, upon which to operate for conversion, or proof of 

 non-conversion, be drawn into rods of about one- fourth of an inch in diame- 

 ter (at each end), but drawn out between these that is, in the middle, into 

 a thin wire tapering gradually towards the thin ends. Fix such a rod in the 

 centre of a glass tube or globe, so shaped and contrived that gaseous mat- 

 ters can be passed into and through it ; connect the obtruding thick ends of 

 the rods with the opposite poles of a voltaic battery powerful enough in 

 its action to raise the thinner portion of the rod to a red or white heat ad 

 libitum, whilst the thicker ends never become so hot as to interfere with the 

 means used to keep them in their position. Fill the tube successively with 

 the following gases, and watch the results : 



Gaseous cyanogen, after a few hours, gives rise to the formation of steel 

 in those portions of the rod that have been fully heated, and this is accom- 

 panied by a deposition of carbon upon the face of the metal. 



Gaseous ammonia per se does not give steel, but produces a curious disin- 

 tegration of the face of the hottest portion of the rod. 



Olefiant gas per se does not give steel, but gives a deposition of carbon on 

 the hotter portions of the wire. But olcfiant gas mixed with ammonia or 

 with nitrogen, does give steel, and so on. 



But without further multiplying examples (and this mode of experiment- 

 ing admits of a great variety), and without attempting at this time to inquire 

 into some of the complicated phenomena they present, one result is ever 

 apparent, namely, the invariable cooperation of both nitrogen and carbon 

 wherever the result is the production of steel. 



The conclusions that, to the writer, appear to be warranted by the pre- 

 vious evidences, are : 



That the substances whose application to pure iron convert it into steel, 

 all contain nitrogen and carbon, or nitrogen has access to the iron during 

 the operation. 



That carbon alone, added or applied to pure iron, does not convert it into 

 steel. 



That nitrogen alone, so added or applied, does not produce steel ; but that 



It is essential that both nitrogen and carbon should be present, and that 

 no case can be adduced of conversion in which both these elements are not 

 present and in contact with the iron. 



That nitrogen as well as carbon exists substantially in steel after its con- 

 version; and such presence is the real cause of the distinctively physical 

 properties of steel and of iron, in which latter these elements do not exist. 



That presumptively, but not demonstratively, the form of combination is 

 not that of cyanogen (though that compound plays an important part in 

 conversion), but is that of a triple alloy of iron, carbon, and nitrogen. 



But that experimental research is yet required to determine the relative 

 proportion of elements when their union gives pure steel. 



What in the chemical history of nitrogen is there that is incompatible 

 with its substantial existence in steel in some form analogous to other 

 combinations we know it to assume under similar conditions with other 

 metals ? Is it under a temperature as high as that needed to melt steel that 

 it combines with carbon to form cyanogen, and then with potassium to 



