"I prepared forty flasks of a capacity of from two hundred and fifty to three hundred cubic centimetres and filled them half full with filtered grape-must, perfectly clear, and which, as is the case of all acidulated liquids that have been boiled for a few seconds, remains uncontaminated although the curved neck of the flask containing them remain constantly open during several months or years.

"In a small quantity of water I washed a part of a bunch of grapes, the grapes and the stalks together, and the stalks separately. This washing was easily done by means of a small badger's-hair brush. The washing-water collected the dust upon the surface of the grapes and the stalks, and it was easily shown under the microscope that this water held in suspension a multitude of minute organisms closely resembling either fungoid spores, or those of alcoholic Yeast, or those of Mycoderma vini, etc. This being done, ten of the forty flasks were preserved for reference; in ten of the remainder, through the straight tube attached to each, some drops of the washing-water were introduced; in a third series of ten flasks a few drops of the same liquid were placed after it had been boiled; and, finally, in the ten remaining flasks were placed some drops of grape-juice taken from the inside of a perfect fruit. In order to carry out this experiment, the straight tube of each flask was drawn out into a fine and firm point in the lamp, and then curved. This fine and closed point was filed round near the end and inserted into the grape while resting upon some hard substance. When the point was felt to touch the support of the grape it was by a slight pressure broken off at the point file mark. Then, if care had been taken to create a slight vacuum in the flask, a drop of the juice of the grape got into it, the filed point was withdrawn, and the aperture immediately closed in the alcohol lamp. This decreased pressure of the atmosphere in the flask was obtained by the following means: After warming the sides of the flask either in the hands or in the lamp-flame, thus causing a small quantity of air to be driven out of the end of the curved neck, this end was closed in the lamp. After the flask was cooled, there was a tendency to suck in the drop of grape-juice in the manner just described.

"The drop of grape-juice which enters into the flask by this suction ordinarily remains in the curved part of the tube, so that to mix it with the must it was necessary to incline the flask so as to bring the must into contact with the juice and then replace the flask in its normal position. The four series of comparative experiments produced the following results:

"The first ten flasks containing the grape-must boiled in pure air did not show the production of any organism. The grape-must could possibly remain in them for an indefinite number of years. Those in the second series, containing the water in which the grapes had been washed separately and together, showed without exception an alcoholic fermentation which in several cases began to appear at the end of forty-eight hours when the experiment took place at ordinary summer temperature. At the same time that the yeast appeared, in the form of white traces, which little by little united themselves in the form of a deposit on the sides of all the flasks, there were seen to form little flakes of Mycellium, often as a single fungoid growth or in combination, these fungoid growths being quite independent of the must or of any alcoholic yeast. Often, also, the Mycoderma vini appeared after some days upon the surface of the liquid. The Vibria and the lactic ferments properly so called did not appear on account of the nature of the liquid.

"The third series of flasks, the washing-water in which had been previously boiled, remained unchanged, as in the first series. Those of the fourth series, in which was the juice of the interior of the grapes, remained equally free from change, although I was not always able, on account of the delicacy of the experiment, to eliminate every chance of error. These experiments cannot leave the least doubt in the mind as to the following facts:

Grape-must, after heating, never ferments on contact with the air, when the air has been deprived of the germs which it ordinarily holds in a state of suspension.

"The boiled grape-must ferments when there is introduced into it a very small quantity of water in which the surface of the grapes or their stalks have been washed.

"The grape-must does not ferment when this washing-water has been boiled and afterwards cooled.

"The grape-must does not ferment when there is added to it a small quantity of the juice of the inside of the grape.

"The yeast, therefore, which causes the fermentation of the grapes in the vintage-tub comes from the outside and not from the inside of the grapes. Thus is destroyed the hypothesis of MM. Trecol and Fremy, who surmised that the albuminous matter transformed itself into yeast on account of the vital germs which were natural to it. With greater reason, therefore, there is no longer any question of the theory of Liebig of the transformation of albuminoid matter into ferments on account of the oxidation."