454 Mr. Graham's Recent Discoveries. 



from the receiver and delivers it at u, a vacuum being ulti- 

 mately produced. 



e represents a bag of silk, varnished with india-rubber, 

 lined with felted carpet. 



After allowing sufficient time for the removal of the original 

 air (representing the capacity of the bag), it will be found that* 

 the air penetrating the bag, and delivered into the tube, e, 

 possesses the power of rekindling a glowing splinter of wood; 

 analysis proving it to contain 40 per cent, of oxygen and 60 

 of nitrogen. 



The constituents of atmospheric air will also pass through 

 rubber into a space containing some other gas, as hydrogen or 

 carbonic acid, at the same relative velocities with which they 

 enter a vacuous space. Later experiments have also proved 

 that air may be changed in composition by escaping under a 

 pressure of two atmospheres through a rubber septum, the 

 proportion of oxygen transmitted being slightly less than in 

 the experiments with the vacuum. 



This penetration of india-rubber by gases is not due to 

 diffusion through actual pores ; if it were, the lighter gases 

 would pass through with the greater velocity ; in the experi- 

 ment with the rubber film, or bag, for instance, diffusion would 

 favour the passage of nitrogen. 



It will be seen, from the table given on page 453, that those 

 gases which penetrate the rubber most readily are those most 

 easily liquified by pressure, and also generally highly soluble 

 in water. 



Mr. Graham considers that the penetration of the gas 

 through rubber is due to its previous absorption as a liquid in 

 the soft, colloid substance of the india-rubber, the transmission 

 being effected by the agency of liquid, and not gaseous 

 diffusion. The rubber being wetted through by the liquified 

 gas, the latter evaporates, and reappears on the other side of 

 the membrane as a true gas. 



There is, moreover, experimental proof of this absorption. 

 When india-rubber is exposed to an atmosphere of carbonic 

 acid gas, it takes up in one hour nearly its own volume, and 

 this gas may be subsequently extracted by the action of the 

 vacuum. Oxygen is twice as soluble in india-rubber as in 

 water, and two and a half times more soluble in rubber than 

 nitrogen is. 



These experiments are of great physiological interest. 

 Respiration is probably due to the liquid diffusion of gases 

 through membranes. The air-bladder of fishes, especially those 

 without a pneumatic duct, must be filled by the same agency. 



In extending the inquiry to the passage of gases through 

 metals, Mr. Graham employed tubes closed at one end, the 



