1825.} On the Specific Gravitif of Hydrogen Gas » 135 



button becomes covered with a layer of steely iron of appre- 

 ciable thickness. 



How does it happen in these experiments, that the oxide of 

 iron is reduced without being in contact with carbon, and even 

 when several centimetres (1 centimetre = 0*39 inch) distant 

 from it ? This is a question which in the present state of our 

 knowledge imperiously demands an answer, and deserves to be 

 considered. We might suppose that the effect is produced by 

 the inflammable vapours from the furnace, which penetrate all 

 porous substances ; but it is easy to satisfy oneself that this is 

 not the efficient cause, at least of the reduction of the oxides of 

 iron into metallic iron. In fact, if we fill a crucible with red 

 oxide of iron, placing a layer of charcoal below it at the bottom 

 of the crucible, or if we place the oxide at bottom and cover 

 it with charcoal ; or lastly, if we introduce charcoal into the 

 centre of a mass of oxide of iron, and heat it for an hour or two, 

 we shall find that metallic iron is formed only in that part of 

 the mass which was next the charcoal, and that there is not the 

 slightest trace of it at the surface of the button in the other 

 parts, although those parts were exposed, like all the rest, to 

 the inflammable gases of the furnace. 



The formation of the forge-scales on the surface of iron is 

 quite as inexplicable as the reduction of the oxides by cemen- 

 tation. The oxidation of hot iron by the air is a gradual process, 

 for the crust of the scales is much thicker on large masses, 

 which require a long time to be heated, than on thin bars or 

 plates, which heat much more quickly : now, as soon as a 

 certain quantity of oxide is formed, it covers the iron like a 

 varnish, and prevents its contact with the air ; it must therefore 

 attract its oxygen through the oxides, just as the oxides attract 

 the carbon through the metallic iron. 



These effects must have certain limits, which it would be 

 important to ascertain, as they may perhaps furnish an expla- 

 nation of the phenomena. 



Article IX. 



lO VA 



On the Specific Gravity of Hydrogen Gas, as^^ modified by the 

 Presence of Moisture! By Mr. Harry Rainy. 



(To the Editors of the Annals of Philosophy.) 



GENTLEMEN, . H J". Glasgow, July 9, 1825. 



Dr. Thomson, in his First Principles of Chemistry, recently 

 published, has adduced various new experiments in proof of the 

 doctrine that the atomic weights of all substances are multiples 

 by integer numbers of tl>e atomic weight of hydrogen. 



