THE AUTOFERMENTATION OF YEAST. 545 



investigated by HENNEBERG (II.) ; according to his view (VI.) it 

 may be assumed that /S'acch. apiculatus does not contain glycogen. 

 Further investigation on the subject of glycogenase would con- 

 stitute a valuable sphere of labour, since all that can be predicted 

 with regard to its still hypothetical existence is the possession of 

 the two properties mentioned. 



A few remarks may be made here respecting the manner in 

 which autofermentation may be influenced by various agencies. 

 We have already seen on p. 172, vol. ii., that, according to 

 Salkowski and Cremer, yeast is affected, by digestion with 

 chloroform water, in such a manner that, whilst the glycogen is 

 split up, the yeast does not undergo autofermentation. C. J. 

 LINTNER (III.) states that the same effect can be produced by the 

 addition of common salt, and he also investigated the action of 

 other saline substances by adding, in all cases, 5 grms. of the salt 

 under examination to each 10 grms. of aspirated bottom yeast 

 rich in glycogen, containing about 25 per cent, of dry substance. 

 Autofermentation was not set up in the samples treated with a 

 chloride of sodium, magnesium, or aluminium, or with ammonium 

 chloride, nitrate, or sulphate. A restrictive influence was pro- 

 duced by the sulphates of manganese and copper, as well as by 

 potassium nitrate. On the other hand, a stimulating effect was 

 brought about by the sulphates of sodium, zinc, magnesium or 

 ferrous oxide, and by monopotassium phosphate. Hence, in 

 certain circumstances, autofermentation may fail to occur in a 

 yeast that is very rich in glycogen, when treated with restraining 

 agents. 



The autofermentation of yeast had a certain practical value 

 for the analytic chemist, namely, when the sugar content of a 

 sample submitted for examination is to be determined by the 

 physiological method (see p. 427, vol. ii.). It is true that methods 

 are available in which the influence of the yeast can be minimised, 

 not only with regard to autofermentation, but also with reference to 

 the other transformations (see pp. 482 and 510, vol. ii.) brought about 

 by yeast, the sterilised solution being inoculated with an im- 

 ponderable trace of yeast, as recommended by ELION (Y.) andBAU 

 (XXXI.). These methods, however, occupy a good deal of time, 

 and though the time question may be neglected in researches of a 

 purely scientific character, it prevents their application in practice, 

 despite their accuracy. For a rapid examination, or commercial 

 analysis, a larger sowing must be used, and in these circumstances 

 the glycogen content and autofermentation of the yeast must be 

 borne in mind, no matter whether the sugar be determined from 

 the resulting carbon dioxide or the alcohol formed. These 

 derivatives are neglected in the method of Elion and Ban, the 

 only determination made being the loss of extract during fer- 

 mentation, the result, supplemented by the cupric reduction 

 and polarimeter tests, giving the actual content of sugar. In 



