DIALTTIC SEPAEATION OF GASES BY COLLOID SEPTA. 423 



Hydrogen 1-93 cub. centim. 



Nitrogen 0-19 „ 



Now, taking the specific gravity of the platinum wire at 21*5, the volume of 201 grms. 

 of metal will be 9*34 cub. centims. Hence one volume of platinum held, the gas being 

 measured cold, 



0-207 vol. hydrogen. 



The platinum did not appear sensibly altered in lustre, or in any other way, by its 

 relation to the hydrogen. 



2. The same piece of platinum wire was drawn out into four times its first length, and 

 the experiment of charging with hydrogen was repeated. The platinum gave up at a 

 red heat, maintained for one hour, 1-8 cub. centim. of gas, of which 1'6 cub. centim. 

 was hydrogen. Here one volume of platinum appears to have held 



0"171 vol. hydrogen. 



The absorption of hydrogen has not been increased by increasing the surface of the 

 metal. 



In two further experiments upon the same platinum wire, the volume of hydrogen 

 retained by one volume of platinum was — 



3. 0-173 cub. centim. hydrogen. 



4. 0-128 



There is an evident tendency of the hydrogen taken up to diminish in quantity. 

 These experiments have the advantage for comparison with the earlier observations on 

 the peneti-ation of a platinum septum by hydrogen gas, that both wire and tube had 

 been drawn out from the same mass of fused metal. No iridium had been added to this 

 platinum, as is sometimes done to increase the elasticity of the metal. The absorption 

 of hydrogen is small, amounting, according to the mean of the four observations, to 17 

 per cent, of the volume of the platinum. At the temperature of a low red heat, when the 

 absorption took place, the gas would be considerably dilated, to the extent of at least 

 three times its volume stated above, or to about 51 per cent., half the volume of the 

 platinum. It is to be considered whether an absorption of half a volume of gas would 

 be sufficient to account for the observed penetration of a septum of metal 1-1 millim. in 

 thickness. The data appear to favour an affirmative conclusion ; but their value cannot 

 be very decidedly estimated. 



It appears necessary to recognize in platinum a new property, a power to absorb 

 hydrogen at a red heat, and to retain that gas at a temperature under redness for an 

 indefinite time. It may be allowable to speak of this as a power to' occlude (to shut 

 up) hydrogen, and the result as the occlusion of hydrogen by platinum. 



The observation was extended to platinum in other conditions of form, but where, it 

 is to be observed, the metal had not been fused, but only welded, and was not of recent 

 manufacture. 



5. Of the grey pulverulent spongy platinum, prepared from the ammonio-chloride, 



WDCCCLXVI. 3 N 



