358 



Prof. H. E. Armstrong and Mr. E. Horton. [Mar. 3, 



effect the decomposition of no less than 13"5 per cent, of the glucoside ' 

 although very active towards methyl-«-glucoside, this extract had no action 

 on amygdalin, a quite unusual circumstance. On repeating the experiment 

 several months later with the same yeast but with an extract prepared by 

 digesting only 15 grammes with 300 c.c. of water, no action was observed 

 either on phaseolunatin or on amygdalin. 



The probability that yeast contains a " j8-glucase " is considerable in view 

 of the fact that cellulose is now known to be a compound of the /3-glucoside 

 class ; cytase (cellase), if not " /8-glucase," is probably a closely allied enzyme. 



It is clear, however, that not only do different yeasts yield different enzymes 

 or different proportions of the several enzymes, but also that age, as well as 

 race and the conditions under which the yeast is dried and extracted, are 

 all factors to be takerf into account as influencing the character of the extract. 



Nature of the Dextrose Residue in FJiaseolunatin. — Dunstan, Henry and Auld 

 give the following description of their experiments made with the object of 

 characterising the glucose produced by the hydrolysis of phaseolunatin : — 



" To a solution of 3 grammes of the glucoside in 20 c.c. of water, 0'5 gramme of the 

 enzyme preparation was added and the mixture kept at 40° in a stoppered bottle. For 

 each observation a few cubic centimetres of the liquid were withdrawn, mixed with a small 



under observation at the ordinary temperature (15°). The following ai'e the results 

 obtained : — 





Glucose. 



Mandelonifcrile 

 glucoside. 



Amygdalin. 





0 



o 







+ 3-89 



-1 -91 



-3 -83 





+ 3-57 



-1 -93 



-3 -84 





+ 3-32 





-3-88 



3 ,. „ 



+ 3-06 



-2 03 



-3 -91 



4 „ „ 



+ 2-89 



-2-10 



-3-94 



5 „ „ „ 



+ 2-73 



-2 15 



-3 -95 



„ f! „ „ „ 



+ 2-63 



-2-18 



-3-98 



„ 7 „ „ „ 



+ 2-63 



-2 -21 



-4-02 



„ « » 



+ 2-48 



-2 -22 



-4-05 



9 ., „ 



+ 2-37 



-2-25 



-4 06 



„ 10 „ „ „ 



+ 2 -33 



-2-31 



-4-08 



From these figures, it is obvious that the t liiinge undergone by glucose is more ra])id 

 than that undergone by mandelonitrilc-glucoside and by amygdalin ; novertholoss, the 

 extent to which racHMnisation of the latter two substances takes place in the time 

 (according to Lowry about 12 minutes in the presence of N/lOO alkali such as we used) 

 required to effect the coni|)lote muta-rotation of glucose is too great to permit the 

 application of E. F. Armstrong's method in the presence of the two cyanophoric 

 glucosides. 



