826 



Professor Sir James Dewar 



in tlie liquid, for water at 15' C. dissolves about one-sixtieth of its 

 volume of air or hydrogen, but the rate of diffusion from air- 

 saturated water is only reduced to one-quarter of the rate in air. 

 Roughly the same relation is shown by hydrogen. The effect of 

 expelling the gases by boiling is very marked, especially in the case 

 of solutions. The effect of glycerin in checking diffusion is also 

 very evident. 



The behaviour of alcohol is quite the opposite. Air goes through 

 the membrane equally rapidly whether alcohol is round it or not. 

 By freeing the alcohol from dissolved gases, the diffusion is checked, 

 as is the case with water ; but in about one hour has regained its 

 initial rate. 



[J. D.] 



