58 PHYSIOLOGY OF THE DOMESTIC ANIMALS. 



It' two different gases are separated by a porous partition, the gases 

 will mix through that partition. The phenomena under such circum- 

 stances are similar to what has been described in the case of liquids 

 under osmosis ; that is, that different gases pass with different degrees 

 of rapidity through the partition, so that the volume of gas increases on 

 one side of the partition and decreases on the other. If an unglazed, 

 porous, earthenware cup (such as is used in a (irove battery) is closed 

 with a cork through which passes a long, vertical, glass tube, whose end 

 dips into a vessel below containing water, and the cup is covered with a 

 bell-jar containing hydrogen or illuminating gas, the hydrogen will pass 

 through the walls of the cup to the inside faster than the air from the 

 inside can diffuse out. The volume of gas in the interior increasing j 

 bubbles of air will escape through the water from the end of the tube. 

 If riow the bell-jar be removed the hydrogen will 

 diffuse out from the cup faster than the air can 

 enter, the volume of gas within the cup will decrease 

 and the water will rise, from atmospheric pressure, 

 within the tube. If oxygen be used within the cup 

 instead of atmospheric air, it will be found that the 

 hydrogen will diffuse four times as fast as the oxygen. 

 The density of hydrogen is 1.; that of oxygen 16.; 

 therefore, the law has been made that the rapidity 

 with which different gases under similar conditions 

 (equal pressures) diffuse through thin, porous par- 

 titions into a vacuum or into other gases is in inverse 

 proportion to the square root of the density of the 

 gases (Graham's law). 



These general facts may be illustrated by another 

 FrcTl^r very simple experiment. If an unglazed earthenware 



cup be closed by a cork in which a water-manometer 

 is inserted and then placed in u, larger glass vessel containing vapor 

 of ether, the air from the inside of the cup will diffuse out faster than 

 the five-times-heavier vapor of ether will diffuse in, and the water in 

 the open arm of the manometer will sink (Fig. 42). 



There is, however, here a marked difference from osmosis, for in the 

 diffusion of gases through inorganic partitions or dry organic membranes 

 the nature of the partition is without influence on the rapidity of diffusion. 

 The rapidity of diffusion depends only on the specific gravity of the gas. 

 10. Absorption of Gases.— Exactly as gases diffuse into spaces 

 already occupied by other gases, so also will they diffuse into the inter- 

 molecular spaces of liquids, without any chemical attraction between the 

 gas and the fluid being essential. If a glass tube, closed at one end, is 

 filled with dry, ammoniaeal gas, its open end immersed in mercury, and 



