PROFESSOR THOMAS GRAHAM's SCIENTIFIC WORK. 211 



The cliaiacteristic property of silver, heated and cooled in difterent 

 atmospheres, proved to be its eapability of absorbing and retaininji', in 

 some cases, as much as seven times its vohime of oxygen — its absorj^ion 

 of hydrogen falling short of a single volume. Some silver-leaf, heated 

 and cooled in ordinary air, and subsequently heated in a vacuum, gave 

 off a mixture of oxygen and nitrogen gases containiijg 85 per cent, of 

 oxygen, or more than four times the proportion contained in the original 

 air. This remarkable property of solid silver to effect the permanent 

 occlusion of oxygen gas, must be distinguished from the not less remark- 

 able and doubtless associated property of melted silver to effect the 

 temporary absorption of a yet larger volume of the same gas : which, 

 on the soliditication of the metal, is discharged with the well-known 

 phenomenon of spitting. 



Iron, though tolerably absorptive of hydrogen, was found to be 

 specially characterized by its absorption of carbonic oxide. What may 

 be called the natural gas of wrought iron, or the gas derived from the 

 forge in which it was heated, proved to consist chiefly of carbonic oxide, 

 and, in difterent experiments, was found to range from 7 to iL'.o times 

 the volume of the metal ; so that, in the course of its i^reparation, iron 

 would appear to occlude upward of seven times its volume of carbonic 

 oxide, and to carry this gas about with it ever after. The absorbability 

 of carbonic oxide by iron has an obviously important bearing on the 

 theory of steel production by cementation. This process would appear 

 to consist in an absorption of carbonic oxide gas into the substance of 

 the iron, and in a subsequent decomposition of the absorbed gas into 

 carbon entering into combination with the metal, so as to etfect its 

 acieration, and carbonic gas discharged from the surface of the metal, so 

 as to produce the well-known appearance of blistering. "Nor is this the 

 only, or even the chief point of interest that was made out with regard to 

 iron ; for the study of the behavior of telluric manutVictured iron 

 naturally led Mr. Graham to the examination of sidereal native iron, 

 that is to say, the iron of meteorites, and with the following result. A 

 portion of meteoric iron, from the Lenarto fall, when heated ;'» vacuo, gave 

 otf 2.85 times its volume of natural gas, of which thepreponderatingcon- 

 stituent, to the extent of 85.7 per cent, of the total (piantity, consisted 

 not of carbonic oxide, but of hydrogen, the carbonic oxide amounting 

 to only 4.5 per cent., and the remaining 9.8 per cent, consisting of nitro- 

 gen. The inference that the meteorite had been, at some time or other, 

 ignited in an atmosphere having hydrogen as its prevailing constituent, 

 seems irresistible; and judging from the volume of gas yielded by the 

 iron, the hydrogen atmosphere in which it was ignited must, in all prob- 

 ability, have been a highly condensed one ; the charge of hydrogen 

 extracted being fully five times as great as it was found possible to im- 

 part to ordinary iron artificially. 



But it was with palladium that Mr. Graham obtained his most extra- 

 ordinary results. This metal he found to have the property of trans- 

 mitting hydrogen with extreme facility, even at temperatures very far 



