60 PASTEUR: THE HISTORY OF A MIND 



mercury, some grapes, the surface of which had been 

 washed many times with hydrogen in order to remove 

 all traces of air from the skins. Then he had crushed 

 them against the top of the tube with the aid of a curved 

 iron rod introduced under the mercury. No fermenta- 

 tion occurred, which might appear very surprising in 

 view of the facility and rapidity with which fermentation 

 ordinarily takes place in the vintage. When it was 

 thoroughly demonstrated that there would not be any, 

 Gay-Lussac brought into contact with the crushed grapes 

 some bubbles of oxygen and saw fermentation begin in 

 a very short time thereafter. From this he concluded 

 that oxygen was necessary to start a fermentation, 

 whatever might be the r61e of the yeast. 



The experiment is accurate, although it does not 

 always succeed. Gay-Lussac tried it twice and failed 

 once. That should have been sufficient to make him 

 reflect on the accuracy of his conclusion, but it was 

 decreed that, in this question, suggestion should play a 

 great rdle. Oxygen was then at the height of its glory, 

 and by opening to it the domain of the fermentations 

 Gay-Lussac not only acted in accordance with the pre- 

 vailing belief, but explained at the same time the pre- 

 serving methods of Appert, who, in heating his bottles 

 and flasks, was seen to be driving out the oxygen, and, 

 in reality, did not leave any of it behind, as experiment 

 showed. Gay-Lussac explained also the very ancient 

 practice of interrupting fermentation by sulphuring casks 

 or the juice of the grapes. Consequently his interpre- 

 tation has entered into people's minds, has remained, 

 and has exercised even on the science of our own time an 

 unquestionable influence. 



Hitherto, the question had been confined to the domain 

 of chemistry. Since Fabroni (1799), the yeast, whatever 

 its r61e, was considered as a gluten, and it did not occur 



