On December 7th, 1874, we took two bottles of must which we had preserved, after Appert’s process, in our laboratory from the beginning of October, 1873. Both of these were covered with dust—the dust that floated about in our laboratory. We decanted them as follows:—One bottle, which we handled without special precautions, we uncorked by means of an ordinary corkscrew, and decanted into another bottle that had been well washed, as bottles are washed when they are to be used subsequently. This bottle was taken from a number that had been standing upside down on a drainer for a fortnight. We took no precaution to remove the dust which covered the exterior of the bottle of must, or to purify the washed bottle. The second bottle of must, on the other hand, was decanted after we had removed the dust that covered it; its cork was cut off close to the string, and the flame of a spirit lamp was passed over the string and the surface of the cork; and, as a final precaution, the corkscrew was passed through the flame. As for the bottle into which we subsequently decanted the must, we first plunged it in a hot-water bath kept at 100° C. (212° F.), then took it into a garden to cool upside down in the open air. After these precautions we removed it, and immediately decanted into it the must from the second bottle. The first bottle showed signs of growths, both on the surface and at the bottom of the must, the day after the operation, and manifested the first symptoms of alcoholic fermentation on December 16th. The contents of the second bottle remained perfectly unchanged after being exposed to the warmth of a stove for several months. Can anything be more conclusive than these facts? They are in perfect keeping with the views that we have recently expressed, and with the principles that we have maintained for nearly twenty years, on the subject of the causes of change in organic liquids. It is by no means our intention to assert that in the atmospheric air there exist no germs of ferment in a state of suspension, as fine dust. Beyond all doubt they do so exist in that state; but, as a rule, in comparatively small number, their abundance or scarcity being dependent upon circumstances which control their multiplication, favouring or restricting it, as we are about to prove. On May 2nd, 1873, we uncorked two ordinary bottles filled with wort, prepared in December, 1872, after Appert’s process. To avoid the causes of error which we have mentioned, we uncorked the bottles in the following manner:—The cork was cut off to the level of the neck; the cork and string were next passed through flame, regardless of burning and charring them; we then gently extracted the cork by means of a corkscrew which also had been passed through the flame. The bottles thus prepared were placed on the table of an underground room, in which we were continually making experiments on alcoholic fermentation.