226 Scientific Litel/igence- [March, 



inches, and the thermometer at 54°. Had the barometer stood 

 at 30 inches, and the thermometer at 60°, the bulk of the gas 

 would have been 14'72 cubic inches. Or (abstracting the azotic 

 gas) 14"5728 cubic inches, I let up into this gas 50 grains of 

 clean iron wire, together with a little water, just sufhcient to 

 moisten the iron. The iron began slowly to rust, and the nitrous 

 gas to diminish in bulk. But as the experiment was made in 

 winter, when the cold was severe, the diminution went on 

 but slowly. After the interval of a month, it seemed to have 

 reached its maximum. I allowed it to remain a fortnight longer, 

 however, and then measured the residual gas. It amounted to 

 nine cubic inches, the barometer standing at 29'94 inches, and 

 the thermometer at 42°. This would have been 9'31'7 cubic 

 inches if the barometer had stood at 30 inches, and the thermo- 

 meter at 60°; or, abstracting the azote, 9* 1698 cubic inches. 

 This residual gas was pure protoxide of azote ; for, being left 

 standing upon water for 24 hours, it was totally absorbed by that 

 liquid, except the 0* 1472 cubic inch of azotic gas with which the 

 nitrous gas had been originally contaminated. 



Now this residual gas, instead of one half, amounts to very 

 near two-thirds of the residual gas. I ask, how this is to be 

 •explained? It is true that in other experiments (made in warm 

 weather) I have seen the bulk of the nitrous gas reduced almost 

 exactly to one half of its original bulk ; but the experiment above 

 related was made with rigid attention to accuracy. Is it not 

 probable from this that the specific gravity of the gaseous residue 

 is not always the same '.' The subject obviously requires further 

 elucidation. 



The common opinion entertained respecting the way in which 

 the oxygen is abstracted from the nitrous gas in this experiment 

 is this. The iron decomposes the water in contact with it, unit- 

 ing with its oxygen, and setting free the hydrogen. The nascent 

 hydrogen acts upon the nitrous gas, abstracting half the oxygeri, 

 and thus converting it into protoxide of azote. Austin and 

 ■Davy, and probably other chemist-s, have shown, that a little 

 ammonia is evolved during this process. Now there is one 

 thing that makes me hesitate about adopting this explanation, 

 simple as it is, as the true one. After the bulk of the gas has 

 decreased so as to reach its maximum, 1 have never observed any 

 increase to take place ; yet if the process had consisted in the 

 decomposition of water by the iron, that decomposition ought to 

 po on as lono; as any water continues in contact with the iron : 

 and after the nitrous gas has been converted into protoxide of 

 azote, the hydrogen, as it does not decompose the protoxide of 

 azute, ought to mix with it, and increase its volume. But I have 

 never found any hydrogen gas mixed Avith the protoxide of azote 

 in this experiment ; and what appears still more conclusive, if 

 possible, I have placed iron wire in contact with a little water in 

 a glass vessel over mercury, and allowed it to remain beside 



