beer with the “high” yeast, which in our experiment of August 2nd, 1873, seemed to have developed in a state of entire purity, this beer when made and stored in cask or bottle could not have failed to undergo a secondary fermentation, in consequence of a development of _saccharomyces pastorianus_. Let us take, as another example of purification of the same kind, the case of the different ferments of the vintage. When must begins to ferment the apiculated ferment invariably appears, and becomes afterwards associated, more or less, with the _saccharomyces pastorianus_, in the presence of which the multiplication of the apiculated ferment soon ceases. _Saccharomyces pastorianus_, in its turn, is gradually displaced by the ferment which we have termed the ordinary ferment of wine, and which Dr. Rees has named _saccharomyces ellipsoïdeus_. On the subject of these changes in the proportion of the ferments of wine, the Note which we published in 1862 in the _Bulletin de la Société chimique_ may be consulted. Now, these various ferments mutually interfere with each other: whereas if _saccharomyces apiculatus_ were there alone it would multiply to a greater extent, and with greater advantage to the fermentation of the must. This result is obtained by filtering the must, as we have already observed. It is evident from what we have just said that the principal part of the deposits of yeast in the sediment of fermented grapes, at the time when the wine is first racked, which in the Jura, is called _l’entonnaison_, is composed of the ordinary ferment of wine, the _saccharomyces ellipsoïdeus_, and that the cells of apiculated ferment are scarcely discoverable with the microscope, being scattered amongst an infinite multitude of other ferments.[110] [Illustration: Fig. 57.]