COMPOSITION OF THE AIR. 415 



haled, as fast as they were produced, died of asphyxia when 

 the quantity of oxygen became reduced to from 3 to 5 per 

 cent. 1 



Following Lavoisier, the Abbe Spallanzani, 2 by researches 

 on a great number of animals of all classes, demonstrated the 

 universal necessity of air, either in a gaseous condition or in 

 solution in liquids, throughout the animal kingdom. 



A few experiments are on record in which the human 

 subject and animals have been made to respire for a time 

 pure oxygen. Though this is the gas which is essential in 

 ordinary respiration, the process being carried on about as 

 well in a mixture of oxygen with hydrogen as with nitrogen, 

 the functions do not seem to be much altered when the pure 

 gas is taken into the lungs. Some authors state that its pro- 

 longed inhalation exaggerates the function for a time, and 

 that inflammation of the lungs and death follow its pro- 

 longed use ; while the experiments of others show that it is 

 harmless. Allen and Pepys confined animals for twenty- 

 four hours in an atmosphere of pure oxygen, without any 

 notable results ; 3 but, as is justly remarked by Longet, these 

 experiments do not show that it would be possible to respire 

 unmixed oxygen indefinitely without inconvenience. As it 

 exists in the air, oxygen is undoubtedly in the best form for 

 the permanent maintenance of the respiratory function. 

 The blood seems to have a certain capacity for the absorp- 

 tion of oxygen, which is not increased when the pure gas is 

 presented. 



The only other gas which has the power of maintaining 

 respiration, even for a time, is nitrous oxide. This is ab- 

 sorbed by the blood-corpuscles with great avidity, and for a 

 time produces an exaggeration of the vital processes, with 

 delirium, etc. properties which have given it the common 



1 BERNARD, op. tit., p. 115. 



3 SPALLANZANI, Memoires sur la Respiration, traduits en Fran$ais d'apres son 

 manuscrit inedit, 1803. 



" LONGET, Traite de Physiologic, Paris, 1861, tome i., p. 458. 



