224 Mr GRAHAM on t/te Law of 



tracted by the fissures, and escapes through them on account of 

 the extreme smallness of its atoms *. 



This explanation is rendered improbable by the circumstance, 

 that hydrogen, of all the gases, was condensed and absorbed with 

 greatest difficulty, and in smallest quantity, by charcoal and the 

 other porous substances, tried by SAUSSURE. And we have no 

 reason to suppose that the particles of hydrogen are smaller than 

 those of the other gases. 



On repeating DOEBEREINER'S experiment, and varying the 

 circumstances, it appeared that hydrogen never escapes out- 

 wards by the fissure without a certain proportion of air returning 

 inwards. In the experiment, however, as originally performed, 

 it is evident, that, as soon as the water rises in the jar above its 

 outer level, air will begin to be forced into the jar mechanically 

 through the fissure, by the pressure of the atmosphere, indepen- 

 dently of what we shall suppose enters by diffusion. But if we 

 press down the jar of hydrogen to a certain depth in the water- 

 trough, so that the level of the water without is kept constantly 

 higher than the level of the water within the jar, then, on the 

 contrary, a portion of the hydrogen will be forced out mechani- 

 cally by the pressure to which the gas is subject. In the last 

 circumstances, however, no air can enter by the fissure, and mix 

 with the hydrogen, except by diffusion, or in exchange for hy- 

 drogen. Now, in a great number of experiments of this kind, 

 the air which entered by diffusion amounted to between one-fifth 

 and one-fourth of the hydrogen, which left the receiver at the 

 same time. But when the circumstances were reversed, and the 

 column of water allowed to rise in the jar above the level of the 

 water-trough, the quantity of air which entered by diffusion was 



* Sur I 1 Action capillaire des Fissures, &c. Annales de Chimie et de Physique, 

 t. 24, pp. 332-334. 1823. 



