On 'the Affaylng of Ores hjy Fujion. 387 



\yy oxygen gas, furnifhed by the deconipofitio'n of attnofphe- 

 ric air : it is an operation which the accurate obferver may 

 daily fee manifefted in almoft every department of the foun- 

 dery and forge. By a juft comprehenfion of it, arid^ proper 

 appreciation of its efFetSsj wfe have a key for the elucidation 

 of the fource and a6lion of many fa6l;s Idng wrappied in myf- 

 tery, or w^hich have beert explained upon principles deftitute 

 of ftability, and unallied to the fundamental operations of 

 truth. 



In place of withdrawing the crucible immediately upon 

 the feparation of the crude iron, were it to remain for an 

 hour longer under an inferior degree of heat, fo as to confo- 

 lidate the metal without again fufing it, then the whole mafS 

 would be found malleable ; the fcoria light and porous. In 

 one of my experiments upon the haematites variety, I ob- 

 tained from I lb. averdupoife, or 7000 Troy grains, an ingot 

 of fine malleable iron weighing 4486 grains j a produce ecjual 

 to 64 per cent. * 



An oxygenating fluX was however iifed, and nearly the 

 whole contents in iron previoufly revived. In experiments 

 without the addition of a flux, the tranfmutation from caft 

 to malleable iron is fhortenedj but the produce is not more 

 than 5-8ths of the intrinfic contents of the ore. Even this 

 procefs may be confiderably fliortencd by removing the cover- 

 ing of vitrified earths floating upon the furface of the metal: 

 thus the fmall portion of the carbonaceous principle conftl- 

 tuting fufibility is almofl inflantly carried off by the com- 

 bination of oxygen 3 the metal lofes its fluidity, becomes 

 thick and clotted, and more fpeedily pafTes into the malle- 



* This miglit be adduced as a proof that crude iron contains more partS 

 ct)ngetiial to malleability than is generally admitted, or than is msniftfted 

 ill the operation of converting it into malleable iron at the forge : the lofs 

 there has already been mentioned to amount to from 30 to ^o per c(n;. of 

 real metal ; in this affay, however, the produce in malleable iron was only 

 4,1; lefs than when accurately aflayed, and rich carbonated crude iron v/a» 

 cbtained j in which, tow, the carbon conftitutcd a part 'J the weight. 



C c ?/ able 



