CHEMISTRY. 181 



the bath during the blow, except in quantity sufficient to furnish 

 a base for the acids present, or if it oxidizes beyond that, it is 

 immediately reduced again, leaving little or no free oxide of iron 

 in the bath. But after the change of flame all this is reversed, 

 and iron oxidizes rapidly and freely, and remains undecornposed, 

 while the residual traces of the other elements as suddenly cease 

 to oxidize rapidly. I freely grant that in referring this back to 

 what is supposed to be a conceded, but unexplained, law, we are 

 merely putting the question in another, a more general, and more 

 abstract shape; still it is, in a qualified sense, an explanation." 

 After a discussion of the effect of the last stage of the process, 

 which is the addition of certain amount of melted spiegeleisen, a 

 pig iron containing from 4 per cent, upwards of manganese and a 

 considerable percentage of carbon, the paper closes with the 

 definition of Bessemer metal as a cast wrought iron, and with a 

 comparison of its merits for various purposes as compared with 

 wrought iron and steel. 



HEAT OF COMBINATION OF BORON AND SILICON WITH 

 CHLORINE AND OXYGEN. 



The products of the oxidation of boron and silicon render the 

 direct determination of the heat evolved by their combustion im- 

 possible. Much interest attaches to such a determination, espe- 

 cially in the case of silicon, which plays such a part in metallurgi- 

 cal operations, as, for example, in the Bessemer process (see p. 

 178). Troost and Haute feuille have presented to the Societe 

 Chimique de Paris a paper on this subject (published in the 

 "Bulletin Mensuel " for March, 1870). They say : 



"In order to obtain these constants it has been necessary to 

 proceed in an indirect manner, and to pass through intermediate 

 combinations. Thus, in the case of silicon, we have been obliged 

 to have recourse to nitre-hydrofluoric acid, the only reagent capa- 

 ble of attacking at ordinary temperatures the various modifica- 

 tions of silicon. Even this reaction, valuable as it is, does not 

 afford the means of determining anything except the differences 

 between the amounts of heat evolved by the combustion of the 

 different modifications of silicon. In order to determine the heat 

 of combination of one modification, amorphous silicon, we were 

 obliged to conduct the experiment in such a manner as to cause 

 the silicon to be attacked by chlorine in the muflle of the calorim- 

 eter. This was effected by mixing with the silicon a small 

 quantity of amorphous boron, which, in combining with the chlo- 

 rine, disengages enough heat to raise a portion of the silicon to a 

 temperature so elevated that the combination with the chlorine 

 begins, and once begun, it continues until the reaction is com- 

 plete. All the experiments were performed with Favre's calo- 

 rimeter." 



The following results were obtained : 



16 



