Mr Graham on Phospkuretted Hydrogen. 91 



continues in general spontaneously inflammable over mercury for 

 forty-eight hours, and sometimes for three or four days, but ceases 

 to be so in a very short time, after the admission of a small pro- 

 portion of air, particularly if the air be added in a gradual man- 

 ner. Thus, if to the gas be passed up one-twentieth part of its 

 bulk of cork or of dry stucco, containing air in its pores, a white 

 smoke appears in the gas, and it ceases to be spontaneously in- 

 flammable in the course of a few minutes. The same mass of 

 stucco, warmed before being passed up into the gas, so as to ex- 

 pel the air it contained, did not produce the same effect. The 

 self-accendible gas always deposits on standing a solid matter con- 

 taining phosphorus, of a lively yellow colour, but in quantity too 

 minute for analysis. This matter is not acted on by any of the 

 ordinary solvents, such as alcohol, ether, alkalies, or muriatic acid, 

 but is destroyed by chlorine-water, and by nitric acid. The pre- 

 cipitation of this matter is most rapid in the case of gas over 

 water, and is indicative of deterioration of the gas. 



2. The self-accendible gas procured from phosphorus, water, 

 and lime, is always mixed with free hydrogen, varying in quan- 

 tity from 25 to 50 per cent. ; while the non-accendible gas from 

 phosphorous acid contains no hydrogen gas, but is pure. Rose 

 concludes that the spontaneous inflammability of the first species 

 cannot depend upon this hydrogen, for the other species is not 

 made self-accendible by the addition to it of any proportion of 

 free hydrogen. On trying the experiment, however, I obtained 

 a different result. A quantity of gas had lost its self-accendibi- 

 lity by standing over water for two or three hours ; to my surprise, 

 the addition to this gas of hydrogen, in any proportion from one- 

 third of a volume to three volumes, restored the self-accendibility 

 of the gas. Spontaneous inflammability was likewise communi- 

 cated, in some cases, to the gas procured from phosphorous acid, 

 merely by adding hydrogen to it. It was early perceived, how- 

 ever, in the course of the investigation, that hydrogen did not 

 uniformly communicate the property in question, and that its in- 



m 2 



