180 ANNUAL OF SCIENTIFIC DISCOVERT. 



2d. That whatever quantities of it are charged into the blast 

 furnace, as fuel, flux, or ore, are almost wholly concentrated into 

 the resulting pig iron ; and, 3d. That no portion is eliminated in 

 the Bessemer converter. How, then, can we hope to make a 

 metal which is good for anything ? 



" Fortunately for our purposes there are ores, fluxes, and fuels, 

 which contain only extremely minute proportions of phosphorus, 

 though I doubt whether there be any such materials absolutely 

 free from it. But there are found those which are sufficiently so 

 for all practical purposes, for there is a limit below which even 

 phosphorus ceases to be injurious to the metal. These furnace 

 materials are somewhat rare, and confined to a few favorable 

 localities, but their existence has been demonstrated fully in 

 Europe, and I am sanguine that they can be found, nay, they 

 are already found, in the United States. 



" With respect to sulphur, the reactions in the Bessemer con- 

 verter do not differ from those of the puddling furnace. It is 

 removed chiefly about the middle of the heat. Although no 

 careful and systematic analyses of the slag have yet been made, 

 it is probable that sulphide of iron will be found there in small 

 quantities, provided the pig metal contain it to the extent of 

 2 or 3 per cent. It is by all means desirable that the pig 

 should be as free from sulphur as from phosphorus. Its effects 

 are always deleterious, although it requires a larger amount of 

 sulphur than phosphorus to cause an equal amount of damage. 

 Like phosphorus, too, it is one of the universally diffused ele- 

 ments accumulated by organic agency, and much more abun- 

 dantly in fuel than in ores. But sulphur is, on the whole, not a 

 very formidable difficulty, because it can be sufficiently removed 

 from most irons. A small quantity of manganese is its best anti- 

 dote, with which it readily combines as a sulphide, with a 

 stronger affinity than for irons. 



" The oxidation of the carbon is the final reaction of the blast. 

 It begins before the silicon is removed and continues with in- 

 creasing vehemence until two-thirds or three-fourths of the time 

 has elapsed, when it begins to show signs of exhaustion. The 

 product is carbonic oxide, with possibly a slight admixture of 

 carbonic acid. The silicon stage very considerably overlaps the 

 carbon stage, indeed, when the percentage of silicon is very 

 high it continues to oxidize in decreasing quantity to the end." 



Alluding to the almost instantaneous " dropping" of the flame 

 at the close of the operation, while a small amount of carbon and 

 other impurities remain and resist further oxidation, Lieut. Dutton 

 says : 



"A plausible explanation is this : When two combustibles are 

 intermixed, like oxygen and hydrogen, or hydrocarbon gas, it is 

 well known that the relative proportion of the two elements in the 

 mixture influences the readiness with which they combine. Thus 

 oxygen and hydrogen cannot combine explosively, unless their 

 proportions lie within certain definite and rather narrow limits. 

 May not the same law hold good in the present case ? It is cer- 

 tain, or nearly certain, that the iron either does not oxidize in 



