DIALYTIC SEPARATION OF GASES BY COLLOID SEPTA. 419 



acid, washed, and dried over chloride of calcium, was then circulated through the outer 

 porcelain tube. The hydrosulphuric acid was nearly all decomposed into sulphur and 

 hydrogen, the latter coming through the platinum at the rate of 9 cub. centims. in five 

 minutes. A trace of hych-osulphuric acid may also have passed through, as the mercury 

 of Sprengel's tube was slightly soiled ; but no indication of this gas could be perceived 

 in the hydrogen collected. It, appears, then, that hydrosulphuric acid is to be classed 

 among the non-penetrating gases. The result appears to be : — 



I.' Gas capable of passing through a septum of platinum 1-1 millim. in thickness at a 



full red heat. 



Hydrogen (211 cub. centims. per hour). 



II. Gases incapable of passing through a septum of fused platinum 1-1 millim. in thick- 

 ness at a full red heat. 



Oxygen .... (not to the extent of 0-2 cub. centim. per hour.) 



Nitrogen .... „ „ 



Chlorine .... „ „ 



Hydrochloric acid . „ . » 



Vapour of water . „ „ 



Carbonic acid . . „ 



Carbonic oxide . . „ 



Marsh gas (C H4) . „ „ 



defiant gas . . . „ „ 



Hydrosulphuric acid „ „ 



Ammonia . . . „ „ 



It remains to be discovered whether a sensible passage of any of these gases could be 

 effected through a platinum septum much reduced in thickness, or through the same 

 septum under the influence of a considerably higher temperature. A fallacious appear- 

 ance of permeation is sometimes occasioned by the escape from the platinum itself of a 

 small quantity of gas, particularly of carbonic oxide and hydrogen, as will immediately 

 appear. The permeation is in consequence never imequivocal for the first hour or two 

 that the platinum septum is heated. 



One of the curious experiments of M. Deville was repeated, in which hydrogen 

 appears to escape from the platinum tube pretty much as the same gas would escape 

 from a graphite diffusiometer — the platinum tube being full of hydrogen, while the 

 annular space between the platinum and outer porcelain tube was occupied by atmo- 

 spheric air. At the maximum temperature the supply of hydrogen to the platinum tube 

 was shut off, as that gas entered at one end of the tube, while the other end of the pla- 

 tinum tube was left in connexion with a barometer-tube dipping into a cistern of mer- 

 cury. Immediately the mercury began to rise in the gauge tube from the passage of 



» 



