18 



A TEXTBOOK OF PHYSIOLOGY 



of each gas is the same throughout and equilibrium is attained. This 

 is termed the process of diffusion of gases ; it is a molecular process, 

 and takes place independently of any movement of gas as a whole by 

 stirring. Diffusion is a relatively slow process, and the mixture of 

 gases is very greatly accelerated by stirring e.g., in the lungs. Ex- 

 perimentally it was found by Graham that the velocity of " effusion," 

 as he termed it, of a given volume of any gas is inversely proportional 

 to the square root of its density. His experiments consisted in 

 determining the times taken for a given volume of various gases, kept 



TIG. 12. HYDROGEN GENERATED FROM A KIFP'S APPARATUS is LIBERATED i> THE 

 NEIGHBOURHOOD OF THE POROUS POT. THE HYDROGEN DIFFUSES INTO THE 

 POT QUICKER THAN THE AIR DIFFUSES OUT, so THAT WATER is FORCED UP THE 



TCBE OUT OF THE WOULFFE'S BOTTLE. 



at constant pressure, to pass through a minute hole in a metal plate 

 into a receiver, which he kept constantly evacuated. This law, 

 governing the rate of diffusion of gases, is well shown by the experi- 

 ment illustrated in Fig. 12. 



Experiments have been made in which the diffusion of a gas is 

 observed through a tube, when the concentration of the gas at one 

 end is kept at zero or at a constant low value. For example, when 

 caustic potash is kept at the bottom of a tall cylinder full of carbon 

 dioxide, there is a constant flow of carbon dioxide to the potash at 

 the \ ottom of the cylinder; the flow is inversely proportional to the 



