The next figure (34) represents the earliest forms of germination of another specimen of _saccharomyces pastorianus_ in wort, after it had been exhausted by four successive growths in sweetened water. We here see the large ferment-form which appears at the commencement of fermentation, in acid fruits, such as cherries and gooseberries (Plate IX.), associated with smaller forms, which follow it and emanate from it, in proportion as the process of budding is repeated. The field was covered with this minute form, and we had to search about considerably before we could find any of the large cells and the long, branching, jointed filaments which we have sketched. The reason of this was, that these large, extended filaments only appear at the beginning, when there is still an abundance of air, giving place, after repeated budding, to minute cells or short filaments, the ever-increasing number of which soon hides the others from sight. [Illustration: Fig. 34.] [Illustration: Fig. 35.] Fig. 35 represents _saccharomyces pastorianus_ again, as it appears after having been exhausted by two years preservation in yeast-water, in contact with pure air. Strange to say, it has lost its elongated appearance, and would appear to have originated from a round ferment. The cells are much exhausted, and most of them seem to have a double border; their interior is very granular and of a yellowish colour. One might readily take the specimen to be a dead old ferment, which, however, it by no means is. Fig. 36 represents the germination of this ferment, which had previously been revived in a flask of wort, at the temperature of the air, in May, 1875. The following are the details of our observation:— We sowed a trace of the exhausted yeast (Fig. 35) in a flask of wort on May 16th. The sketch (Fig. 36) was made on May 19th, but on the 18th there was a sensible revival. It will be seen how much the little ferment had developed in the course of three days from the time when the process commenced. If we had waited a few days longer before taking our sample, we should probably have had difficulty in finding any cells or filaments of the large ferment form, as there would have been so few of them in comparison with the others. [Illustration: Fig. 36.]