PROPERTIES OF EXPRESSED YEAST JUICE. 465 



of fermentative power can be detected at the end of a year ; and 

 these results were confirmed by R. ALBERT (I.), The retention 

 of fermentative power by sterile permanent yeast will be 

 mentioned on p. 476. 



A similar autofermentation to that occurring in living yeast 

 (see chapter Ixv.), is also noticed in expressed yeast juice. It is 

 attributed to the glycogen content of the yeast (see p. 171, 

 vol. ii.). In conformity with WILL'S (XX.) discovery of the low 

 percentage of glycogen in Munich pressed yeast, the auto- 

 fermentation in Munich yeast juice is slight, amounting 

 according to BUCHNER and RAPP (VII.) to not more than 

 corresponds to 0.45 grm. of carbon dioxide per 100 c.c. of the 

 juice. According to E. BUCHNER (VII.) it is greater in Berlin 

 yeast juice, and corresponds to 0.40-1.10 grms. of carbon dioxide 

 per 100 c.c. of juice. The maximum value exceeding even the 

 fermentation in presence of saccharose was that obtained by 

 MACFADYEN, MORRIS and ROWLAND (I.) with juice from top- 

 fermentation yeast, namely, 65-900 c.c. of gas per 100 c.c. of the 

 juice; unfortunately, however, the cause of this unusually 

 extensive autofermentation was not further investigated. The 

 result is out of harmony with those of E. BUCHNER (VII.), and is 

 probably attributable to the presence of living bacteria. In a 

 subsequent investigation conducted by HARDEN and YOUNG (I.), 

 the auto-fermentation, though greater than in Buchner's experi- 

 ments, did riot exhibit such considerable differences as in those of 

 Macfadyen, Morris and Rowland. 



Divergent results have also been obtained by the various 

 workers on filtering the juice through a bacterium filter. 

 Whereas BUCHNER and RAPP (I.), and also MACFADYEN, MORRIS 

 and ROWLAND (I.), observed merely a diminution in the fer- 

 mentative power of the juice when forced through a Cham- 

 berland filter especially in comparing the first and subsequent 

 fractions WROBLEWSKI (III.) found that this function of 

 the juice was entirely destroyed by the treatment in question. 

 Similar results were obtained by STAVENHAGEN (I.) after filtering 

 the juice through a Kitasato filter. This effect is probably 

 explained by the circumstance that bacteria filters do not permit 

 the transfusion of proteids to more than a slight extent, if at all 

 (see vol. i., pp. 99, 100). 



Hence the filtration of expressed yeast juice through a very 

 fine filter may result in a loss of proteids, and, therefore, of the 

 fermentative enzyme. For this reason it is important to bear 

 the initial fermentative power of the juice in mind during 

 experiments of this kind ; for, since a decrease of this property 

 must be expected from filtration, the initial power must be very 

 high if the juice is to exert any fermentative action at all 

 afterwards. This is the sole explanation of the unfavourable 

 result obtained by Stavenhagen. It should also be mentioned 



