350 LOUIS PASTEUR 



inches) of gas, containing ten per cent, of hydrogen. On 

 the 2nd we began the study of the action of air on the 

 vibrios of this fermentation. To do this we cut off the 

 delivery-tube on a level with its point of junction to the 

 flask, then with a 50 cc. pipette we took out that quantity 

 (i^4 fl. oz.) of liquid which was, of course, replaced at once 

 by air. We then reversed the flask with the opening under 

 the mercury, and shook it every ten minutes for more than 

 an hour. Wishing to make sure, to begin with, that the 

 oxygen had been absorbed we connected under the mercury 

 the beak of the flask by means of a thin india-rubber tube 

 filled with water, with a small flask, the neck of which had 

 been drawn out and was filled with water; we then raised 

 the large flask with the smaller kept above it. A Mohr's 

 clip, which closed the india-rubber tube, and which we then 

 opened, permitted the water contained in the small flask to 

 pass into the large one, whilst the gas, on the contrary, 

 passed upwards from the large flask into the small one. We 

 analyzed the gas immediately, and found that, allowing for 

 the carbonic acid and hydrogen, it did not contain more than 

 14.2 per cent, of oxygen, which corresponds to an absorption 

 of 6.6 cc., or of 3.3 cc. (0.2 cubic inch) of oxygen for the 

 50 cc. (3.05 cubic inches) of air employed. Lastly, we again 

 established connection by an india-rubber tube between the 

 flasks, after having seen by microscopical examination that 

 the movements of the vibrios were very languid. Fermenta- 

 tion had become less vigorous without having actually 

 ceased, no doubt because some portions of the liquid had not 

 been brought into contact with the atmospheric oxygen, in 

 spite of the prolonged shaking that the flask had undergone 

 after the introduction of the air. Whatever the cause might 

 have been, the significance of the phenomenon is not doubt- 

 ful. To assure ourselves further of the effect of air on the 

 vibrios, we half filled two test tubes with the fermenting 

 liquid taken from another fermentation which had also at- 

 tained its maximum of intensity, into one of which we 

 passed a current of air, into the other carbonic acid gas. In 

 the course of half an hour, all the vibrios in the aerated tube 

 were dead, or at least motionless, and fermentation had 

 ceased. In the other tube, after three hours' exposure to the 



me 



