A powerful digestive enzyme of the type of trypsin is also present, so that when the juice is preserved its albumin undergoes digestion at a rate which depends on the temperature [Hahn, 1898; Geret and Hahn, 1898, 1, 2; 1900; Buchner, E. and H., and Hahn, 1903, pp. 287-340], and is converted into a mixture of bases and amino-acids. After about six days at 37°, or 10 to 14 days at the ordinary temperature, the digestion is so complete that no coagulation occurs when the juice is boiled. As this proteoclastic enzyme, like the alcoholic enzyme, cannot be extracted from the living cells, it is termed yeast endotrypsin or endotryptase. Fresh yeast-juice produces a slow fermentation of sugar, which lasts for forty-eight to ninety-six hours at 25° to 30°, about a week at the ordinary temperature, and then ceases, owing, not to exhaustion of the sugar, but to the disappearance of the fermenting agent. When the juice is preserved or incubated in the absence of a fermentable sugar this disappearance occurs considerably sooner, so that even after standing for a single day at room temperature, or two days at 0°, no fermentation may occur when sugar is added. The reason for this behaviour has not been definitely ascertained. As will be seen later on (p. 64) the phenomenon is a complex one, but the disappearance of the enzyme was originally ascribed by Buchner to the digestive action upon it of the endotrypsin of the juice [1897, 2], and no better explanation has yet been found. Confirmation of this view is afforded by the fact that the addition of a tryptic enzyme of animal origin greatly hastens the disappearance of the alcoholic enzyme [Buchner, E. and H., and Hahn, 1903, p. 126], and that some substances which hinder the tryptic action favour fermentation [Harden, 1903]. The amount of fermentation produced is almost unaffected by the presence of such antiseptics as chloroform or toluene, although some others, such as arsenites and fluorides, decrease it when added in comparatively high concentrations, and it is only slightly diminished by dilution with three or four volumes of sugar solution, somewhat more considerably by dilution with water. When it is filtered through a Chamberland filter the first portions of the filtrate are capable of bringing about fermentation, but the fermenting power diminishes in the succeeding portions and finally disappears. The juice can be spun in a centrifugal machine without being in any way altered, and no separation into more or less active layers takes place under these conditions. [p021] The amorphous powder obtained by drying the precipitate produced when the juice is added to a mixture of alcohol and ether is also capable of producing fermentation, and the process of precipitation may be repeated without seriously diminishing the fermenting power of the product.