Dialytic Separation of Gases by Colloid Septa. 519 



of selection and absorption of alcohol in preference to water. 

 30 grms. of the sponge were left in contact with 9*5 cub. cen- 

 tims. of dilute alcohol of specific gravity 0*893, for fifty-one 

 hours, in a sealed tube. The supernatant liquid now drawn off 

 to the extent of 3*9 cub. centims. was of specific gravity 0*901, 

 while the portion retained by the palladium was found when 

 distilled to be of specific gravity 0*885, or it was sensibly con- 

 centrated. This chemical action of palladium sponge was more 

 than once verified. Platinum sponge, on the other hand, exhi- 

 bited no indication of a similar separating-power ; nor did the 

 sponge of iron reduced by hydrogen from the oxide. 



8. Connected, it may be, with this chemico-molecular action 

 of palladium is the variable absorptive power for different liquids 

 exhibited by palladium-foil. Immersed in various liquids for an 

 hour, and afterwards dried by pressure for a few seconds between 

 folds of blotting-paper, a quantity of palladium- foil represented 

 by 1000 was found to retain in its pores — 



Of Water 1*18 part. 



Of Alcohol (0*802) ... 5*5 parts. 



Of Ether 1*7 part. . 



Of Acetone (0*794) . . 0*54 „ 



Of Glycerine 4*5 parts. 



Of benzol 3*5 „ 



Of Oil of sweet almonds .18*1 „ 



Of Castor-oil .... 10*2 „ 



The superior penetrativeness of alcohol over water is well 

 marked ; capillary action appears to merge into a chemical affi- 

 nity. Liquid hydrogen would also appear as highly absorbable 

 by palladium-foil. It would appear also to be separable from 

 other gases (or liquids), as alcohol is from water, by the palladium- 

 pores. 



Alloy of 5 palladium and 4 silver. — The power to absorb hy- 

 drogen appears to extend to this alloy of palladium. A plate of 

 the alloy, about 180 millims. in length, 31 millims. in width, 

 and weighing 74*3 grms., was bent, so as to be able to enter a 

 wide porcelain tube that could be exhausted of air when required. 

 The volume of the palladium alloy was 621 cub. centims. The 

 plate of metal being placed in the porcelain tube, had hydrogen 

 gas passed over it at a low red heat for one hour, and was then 

 allowed to cool slowly in the same gas. Taken out and examined, 

 the metal was not visibly altered. For the extrication of gas the 

 metal was distilled in the porcelain tube heated by jets of gas, 

 aid connected with the Sprengel pump, as usual. In seven 

 minutes after the gas-furnace was lit, 24 cub. centims. of gas 

 eame off; in ten minutes more, 80*71 cub. centims.; and in 



