90 SCIENTIFIC AMUSEMENTS. 



pumped out is much richer in oxygen than the original outside 

 atmosphere ; instead of containing only one-fifth of its bulk of pure 

 oxygen, it will contain about two-fifths. This action is obviously 

 analogous to the more easy passage of water than of alcohol 

 through a bladder (Expt. 40), or to the alteration produced in the 

 composition of a given gas contained in a soap bubble, when 

 allowed to stand some time, the gas dissolving in the watery part 

 of the soap film passing through, and evaporating again into the 

 outside air, whilst air passes inwards by the reverse operation 

 (Expt. 79) ; or the similar action of a moist bladder containing 

 air when placed in a gas of carbon dioxide (Expt. 80). 



If a jar of hydrogen gas be tied over, jam-pot fashion, with a thin 

 sheet of india-rubber, the hydrogen will slowly pass outwards, and 

 the external air inwards through the rubber ; the former action 

 goes on more rapidly than the latter, so that the total gaseous 

 contents of the jar are lessened ; accordingly, if the temperature 

 remains the same so as not to affect the result by altering the 

 volume of the gas in the jar, the rubber cover will be gradually 

 pressed inwards and become concave. Conversely, if a jar filled 

 with air and similarly tied over be allowed to stand some days 

 inside a larger jar filled with hydrogen, the latter gas will pass 

 inwards more rapidly than the air will pass outwards, so that the 

 rubber will become convex, bulging outwards. 



Platinum and palladium, and to some extent iron, more especially 

 when considerably heated, possess a similar power of allowing 

 hydrogen to pass through walls of these metals more readily than 

 most other gases; red-hot iron is also permeable to carbon monoxide 

 to some considerable extent. This property is well illustrated in the 

 case of coal gas (a complex mixture of several gases, of which 

 hydrogen is one main constituent), by enclosing a long tube of 

 platinum or palladium inside a glass tube, which thus serves as a 

 jacket, just as in a Liebig's condenser (fig. 23). The central part 

 of the compound tube is heated, and a current of coal gas led 

 through the jacket, whilst the inside of the metal tube (stopped 

 up at the far end) is connected with a good air-pump. Under 

 these circumstances, the hydrogen of the coal gas passes pretty freely 

 through the metal, and is extracted in a nearly pure state by the 

 pump, whilst the other gases do not pass through at all, or at least 

 only to an inconsiderable extent. The action may to some extent 

 be compared with that of a wet paper filter into which a mixture 

 of oil and water is placed (fig. 36) : if the filter be wet with water, 

 only water passes through, and oil remains behind ; whilst if the 

 paper be previously impregnated with oil, only oil passes through, 

 the water being now left behind (vide Expt. 84). 



