law; on the normal pressure being restord, the gases with which it was supersaturated pass into the free state. It is like drawing the cork of a bottle of beer. The oxygen combines on the spot, but the nitrogen is at once set free, and carries with it carbonic acid in becoming disengaged. Death is explained by the arrest of the circulation.
But it must not be supposed that the action of compressed air is harmless. If we subject a sparrow to a pressure of twenty atmospheres, it will, after a few minutes, be seized with tremors, increasing to most violent convulsions—convulsions stronger than those of tetanus or of strychnine-poisoning—and the bird soon dies. These terrible symptoms are not the result of compression, as I have been able to prove by two experiments. In the first place, they can be produced at the pressure of five atmospheres, provided pure oxygen be used instead of air, which latter has no special effect at this pressure. Secondly, they do not make their appearance if the air subjected to twenty atmospheres' pressure is very poor in oxygen.
Thus it is the oxygen that is to blame. Oxygen at too high a degree of tension destroys animal life. Long I hesitated to characterize as a poison the "nursing father" of everything that lives, but there was no help for it. Oxygen, which gives us life, slays also, when administered in too strong a dose. I have had to study thoroughly this paradoxal poison to determine the different effects of varying doses, and its action upon our tissues.
Here a new surprise awaited me. Having seen a sparrow killed by oxygen, I supposed that this agent must have accelerated organic combustion, thus consuming all the material which goes to maintain the animal heat. But great was my astonishment when the thermometer indicated in animals laboring under strong convulsions a fall of several degrees in the temperature. The analysis of other phenomena confirmed this first observation, and led me to the strange conclusion that oxygen in excess kills by interfering with, arresting, the intra-organic oxydation.
The effects of this powerful agent begin to be distinctly felt at the pressure of about five atmospheres. Perhaps they might be noticed at a lower pressure, and I am inclined to attribute to this cause the unfavorable symptoms presented by workmen who have spent several months in compressed air; but this is a complex problem. In any case, if the necessities of industry subject men to pressures higher than six atmospheres, they will be in danger not only at the instant of decompression, but even from the effects of the compression.
Oxygen at a high tension kills not only the higher animals: it acts alike on vertebrates and invertebrates, animals aerial and aquatic, plants and animals, big and little, even microscopic organisms. Its effects upon the latter are highly interesting.
From the admirable researches of Pasteur it appears that the phenomena of fermentation are of two kinds. One set of phenomena