506 Mr. T. Graham on the Absorption and 



Passage of hydrogen gas in one minute through a septum of 

 1 square metre, — 



Through rubber 0014 millim. in thickness, 127*2 cub. cen- 

 tims. at 20° C; 



Through platinum 1*1 millim. in thickness, 489*2 cub. cen- 

 tims. at bright red heat. 



If the permeation of hydrogen is due to the same agency in 

 both septa, can the vast superiority of the platinum septum be 

 connected with its greatly higher temperature ? 



It was interesting now to turn from hydrogen to the passage 

 of other gases through heated platinum. The experiments were 

 all made in the same way, and at a full red heat. The tempera- 

 ture, it will be observed, was short of that at which the elements 

 of water and carbonic acid are partially dissociated. 



Oxygen and Nitrogen. — Atmospheric air, which may be taken 

 to represent both of these gases, was now allowed to flow through 

 the annular space between the tubes, the interior platinum tube 

 being kept vacuous as usual. In one hour the gas collected 

 by the constant action of a Sprengel pump amounted only to 0*3 

 cub. centim. Hydrogen in the same time would have given 211 

 cub. centims. It is very doubtful, too, whether the trifling frac- 

 tion of a centimetre of gas collected had all passed through the 

 platinum ; a part (or the whole of it) may have entered by the 

 joints of the apparatus. Platinum, then, cannot be said to be 

 sensibly permeable to either oxygen or nitrogen, even at a full 

 red heat. 



Carbonic acid. — This gas was supplied from a bottle contain- 

 ing marble, by the action of pure hydrochloric acid, the gas being 

 afterwards washed with water and dried by sulphuric acid in its 

 way to the exterior porcelain tube. In one hour the interior 

 platinum tube yielded only three-tenths of a cubic centimetre of 

 gas, of which, again, only an indeterminate small portion was 

 condensed by baryta-water and appeared to be carbonic acid. 

 The passage of carbonic acid is therefore incalculably small at a 

 full red heat. 



Chlorine. — This gas, evolved slowly from a glass flask containing 

 peroxide of manganese and hydrochloric acid, was washed by 

 water, dried by sulphuric acid, and thrown as usual into the 

 porcelain tube so as to occupy the annular space between the 

 two tubes. A small tube containing slaked lime was interposed 

 between the end of the platinum tube and the Sprengel pump, 

 so as to absorb the chlorine, if any came through the substance 

 of the platinum. After the tube had been heated for an hour, 

 the lime was examined for chlorine, but did not contain a trace of 

 that substance. A minute quantity of gas, probably air, amount- 

 ing to 0*15 cub. centim., was collected during the time. Plati- 



