The following observations will afford more than sufficient proof of this statement, being, after all, but an easy method of carrying out Gay-Lussac’s experiment, without having recourse to the use of mercury. It may already be inferred from the preceding facts that there must be, in the course of the year, between the end of winter and autumn, a period when the vegetation of the cellules from which yeast proceeds undergoes a revival. When does this period occur? In other words, how long after winter does sterility of the plant continue, until it is again capable of yielding ferment? To ascertain this, we conducted numerous experiments during the summer and autumn of 1875 and the winter of 1876. Having to conduct them in a vine-growing country—in the vineyards of Arbois, Franche-Comté—at a distance from our laboratory, we were compelled to adopt a simple form of apparatus for our experiments, which, besides being very convenient, was at the same time sufficiently exact for the object we had in view. [Illustration: Fig. 32.] Into common test-tubes we poured some preserved must; we then boiled it, with the object of destroying all the germs that it might contain, and then, having passed the flame of a spirit lamp over the upper sides of the tubes, we closed them with corks which had been held in the flame until they began to carbonize (Fig. 32). Having provided ourselves with a series of tubes prepared in this manner, we carried them to a vine, and there dropped into some of them grapes, into others bunches, from which we had taken all the grapes, by cutting their peduncles; into others, fragments of leaves or the wood of the branches. The corks were again passed through the flame and replaced successively in each tube. Some of the grapes we dropped in whole, some we crushed at the bottom of the tubes with an iron rod that had previously been passed through the flame; others, again, at the same moment that we introduced them into the tubes, were cut open with scissors, likewise passed previously through the flame, so that a portion of their interior juice might mix with the must in the tube. Our experiments gave the following results:—As long as the grapes were green, about the end of July and during the first fortnight of August, we obtained no fermentation in our must. Between the 20th and 25th of August a few tubes underwent fermentation, by the action of the little apiculated ferment; and in the course of September the number of tubes that fermented increased progressively. In each series of tubes, however, we always found a few in which there was a complete absence of fermentation.