The cessation of fermentation in any particular mixture of enzyme and co-enzyme may, therefore, be due to the disappearance of either of these factors from the liquid. If the amount of co-enzyme present be relatively small it is the first to disappear, and fermentation can then only be renewed by the addition of a further quantity, whilst the addition of more enzyme produces no effect. If, on the other hand, the amount of co-enzyme be relatively large, the inverse is true; the enzyme is the first to disappear, and fermentation can only be renewed by the addition of more enzyme, a further quantity of co-enzyme producing no effect. It has, moreover, been found that the co-enzyme, like the enzyme, disappears more rapidly in the absence of glucose than in its presence, incubation at 25° for two days being as a rule sufficient to remove all the co-enzyme from yeast-juice from top yeasts in the absence of sugar, whilst in the presence of fermentable sugar co-enzyme may still be detected at the end of four days. In all the experiments carried out by Harden and Young with juice from English top yeast it was found that when a mixture of the juice with glucose was incubated until fermentation had ceased, the further addition of co-enzyme in the form of boiled juice did not cause any renewal of the action; in other words, the whole of the enzyme had disappeared. On the other hand, Buchner and Klatte [1908], working with juice and zymin prepared from bottom yeast, observed the extremely interesting fact that after the cessation of fermentation the addition of an equal volume of boiled juice caused a renewed decomposition of sugar, and that the processes of incubation until no further evolution of gas occurred and re-excitation of fermentation by the boiled juice could be repeated as many as six times. Thus in one experiment the duration of the fermentation was extended from three to a total of twenty-four days, and the total gas evolved from 0·73 gram to 2·19 grams. The phenomenon has been found to be common to yeast from Munich and [p065] from Berlin as well as to zymin and maceration extract, and it was further observed that the boiled juice from one yeast could regenerate the juice from another, although the quantitative relations were different.