
On the Probable Existence of Emulsin in Yeast. 569 
has been employed for the isolation of mandelonitrile glucoside,* and of 
dhurrin and phaseolunatin.f 
With a view to devising a general process for the isolation of cyanogenetic 
glucosides we have, at Professor Dunstan’s suggestion, investigated more 
thoroughly the properties of some of the known glucosides of this type, and 
the present paper contains an account of a number of results obtained in 
attempting to remove dextrose from mixtures of this sugar with cyanogenetic 
glucosides, by fermentation with yeast. 
Action of Yeast on Amygdalin. 
As a preliminary experiment, a solution of a mixture of equal quanti- 
ties of amygdalin and dextrose in water was mixed with a small quantity 
of ordinary pressed yeast and allowed to stand in a warm place. Fermen- 
tation took place, carbon dioxide was evolved, and after several days a 
distinct odour of oil of bitter almonds was observed. This unexpected 
decomposition of the glucoside rendered necessary confirmatory experiments 
with amygdalin alone. 
Amyedalin (2 grammes) was dissolved in 100 cc. of water and about 
6 grammes of ordinary pressed yeast added, together with a few drops of 
toluene to render the mixture antiseptic. The experimental flask was plugged 
with cotton wool and kept, together with a control flask containing a solution 
of amygdalin in water without yeast, at 40°. In the flask to which yeast had 
been added, the odour of benzaldehyde was observed after three days. This 
rapidly increased in intensity. After standing for two days more the mixture 
was distilled until free from hydrocyanic acid and the distillate, previously 
rendered slightly alkaline with potash, was extracted with ether. The oily 
residue left on distilling off the solvent had a strong odour of benzaldehyde, and 
the presence of this substance was proved by its conversion into dibenzyli- 
deneacetone (melting point, 112°), by condensation with acetone in presence 
of potash, and into benzaldehyde phenylhydrazone by the action of phenyl- 
hydrazine. 
The aqueous solution left after extraction with ether was diluted to a 
known volume with water, and a portion examined for prussic acid with 
positive results. An aliquot part of the whole was then titrated with standard 
silver nitrate solution by Liebig’s method, and by this means it was found 
that 33 per cent. of the amygdalin originally present in the solution had 
been decomposed by the yeast. The control flask was similarly examined, 
and it was found that in it no decomposition had occurred. A second experi- 
* Fischer, ‘ Berichte,’ 1895, vol. 28, p. 1509. 
+ Dunstan and Henry, Joc. cit. 
