116 REPORT— 1853. 



indicate up to the tenth and twelfth meltings ; but it is probably not more 

 than what might naturally be expected from the number of meltings and the 

 quantity of limestone used each time as a flux. This flux, when in contact 

 with the iron at a high temperature, would part with a portion of its silica, 

 or, what is equally probable, the iron would take up its equivalent at each 

 melting, and thus bear out the fact recorded in the table of the increase and 

 relative importance of this constituent. 



As respects the sulphur, the surprise is that the increase is not greater, 

 and that more particularly when the quantity contained in the fuel used is 

 considered. In this constituent we are led to the conclusion that a gradual 

 absorption from the coke must have taken place from the commencement to 

 the end of the process, and in this view of the question the quantity taken 

 up by the iron at the diflPerent meltings should have been nearly the same. 

 This is, however, not the case, as we find the tenth melting of a purer quality 

 by '34! than the eighth, and '^Q less than those of the eighteenth melting. 



In the relative increase of the quantity of carbon * there is not that dis- 

 crepancy, as in the absorption of this constituent there is not a uniform but 

 a variable increase; and from the first to the last melting there is an increase 

 from 2*76 to 3'75, or nearly one per cent. 



The cliemical properties of these difl!'erent meltings are somewhat peculiar 

 in character, and appear to be entitled to further investigation, and that by 

 abler and more intelligent heads tlian my own. On some future occasion I 

 hope to induce some of my chemical friends to take up the subject, and 

 nothing will give me greater pleasure than to furnish the necessary facilities 

 for such an inquiry. 



* Mr. Calvert states, in his note attached to the analyses, that the quantity of carbon 

 contained in the specimens was determined in the usual way ; but the process adopted, 

 although the best in use, was not calculated, in his opinion, to enable him to state the real 

 amount of carbon in each specimen. 



