Death in Closed Vessels 539 



still more remarkable is that I could not use the precautions the 

 importance of which I stressed above, and that the decompression 

 was therefore always very sudden. 



Applying to oxygen the reasoning used on page 528 and fixing 

 as limits of minimum tension of oxygen compatible with life the 

 numbers 3 and 4.2, we find for the lowest barometric pressures to 

 which it would be possible theoretically, in pure oxygen, with all 

 necessary slowness and precautions, to bring sparrows without 



x 

 killing them the numbers taken from the equations 100 x — = 3 



76 

 x 

 and 100 x — = 4.2. Whence x = 2.3 cm. and x = 3.2 cm. It is 



76 

 evident that in practice one cannot go so low. 



The ultimate phenomenon, that is, death, is not the only one 

 whose barometric limit varies with the oxygen percentage of the 

 medium. The other disturbances, uneasiness, cessation of move- 

 ments, vomiting, general weakness, are in the same category. It 

 has always been easy to prove that the bird which seemed sick at 

 first diminution of pressure, when it reached 40 cm. for example, 

 gave no sign of uneasiness when, after admitting oxygen, I de- 

 creased the pressure again and reached the same level. I had to 

 go farther, to 50 cm. for example, to get the same morbid 

 phenomena. 



The experiments with super-oxygenated air have therefore com- 

 pletely proved what the experiments with ordinary air showed to 

 be certain. It would have been possible to get a counter-verifica- 

 tion from experiments in which we would have used air poor in 

 oxygen. I could give with details some data of this sort; but the 

 proof must already be thoroughly convincing in the reader's mind, 

 and I shall merely say that with air containing only 10.2 per cent 

 of oxygen, I could not reach a pressure lower than 28 cm., the 

 oxygen tension then being 3.7. 



It is then established that either in a closed vessel by respiratory 

 depletion, or in a current of air, death occurs in rarefied air in 

 consequence of a lessening of the tension of the ambient oxygen. 

 Diminution of barometric pressure is only one method of obtaining 

 this insufficient tension. But there is a second method which con- 

 sists of lowering the percentage of oxygen; we need only consider 



2 x P 



the equation mentioned s -> often already = 3.6. 



76 



