PASTEUR, KOCH, AND OTHERS 287 



He planned and executed one experiment which he sup- 

 posed was conclusive. In introducing it he said: "The 

 opponents of spontaneous generation assert that the germs of 

 microscopic organisms exist in the air, which transports them 

 to a distance. What, then, will these opponents say if I 

 succeed in introducing the generation of living organism-, 

 while substituting artificial air for that of the atmosphere? ' 



He filled a flask with boiling water and sealed it with great 

 care. This he inverted over a bath of mercury, thrusting 

 the neck of the bottle into the mercury. When the water 

 was cooled, he opened the neck of the bottle, still under the 

 mercury, and connected it with a chemical retort containing 

 the constituents for the liberation of oxygen. By heating 

 the retort, oxygen was driven off from the chemical salts 

 contained in it, and being a gas, the oxygen passed through 

 the connecting tube and bubbled up through the water of 

 the bottle, accumulating at the upper surface, and by pressure 

 forcing water out of the bottle. After the bottle was about 

 half filled with oxygen imprisoned above the water, Pouch et 

 took a pinch of hay that had been heated to a high tempera- 

 ture in an oven, and with a pair of sterilized forceps pushed 

 it underneath the mercury and into the mouth of the bottle, 

 where the hay floated into the water and distributed itself. 



He thus produced a hay infusion in contact with pure oxy- 

 gen, and after a few days this hay infusion was seen to be cloudy 

 and turbid. It was, in fact, swarming with micro-organisms. 

 Pouchet pointed with triumphant spirit to the apparently 

 rigorous way in which his experiment had been carried on: 

 "Where," said he, "does this life come from? It can not 

 come from the water which had been boiled, destroying all 

 living germs that may have existed in it. It can nol come 

 from the oxygen which was produced at the temperature 

 of incandescence. It can not have been carried in the 1k>v, 

 which had been heated for a long period before being intro* 



