1906.] 



Cyanogenesis in Plants. 



157 



solution was evaporated to dryness and the residue set aside so that 

 crystallisation could occur, but it was only after this tedious process of 

 dissolution in alcohol, precipitation by excess of ether, and evaporation to a 

 syrup had been repeated live times that a crystallisable residue was obtained. 



The difficulty of separating the glucoside from these residues appears to be 

 due to the saccharine and extractive matters present, and it is only by the 

 practically complete removal of these impurities by precipitation from 

 alcoholic solution with excess of ether that the glucoside can be induced to 

 crystallise. This process is both tedious and wasteful, the greater part of the 

 glucoside being included in the several uncrystallisable fractions obtained 

 when the solutions in alcohol are poured into excess of ether. 



The crystalline residue eventually obtained was recrystallised from alcohol 

 until colourless and of constant melting point. It crystallised in the 

 spreading rosettes of colourless needles which are characteristic of phaseo- 

 lunatin, had the same cool, bitter taste, and, like it, was readily soluble in 

 water, less so in alcohol, and almost insoluble in dry ether. It melted 

 at 138° (corr.). A mixture of phaseolunatin, prepared from the seeds oE 

 Phaseolus lunatus, and the cassava glucoside also melts at this temperature. 

 There can be no doubt, therefore, that the glucoside of cassava is identical 

 with phaseolunatin, which we have shown to be a dextrose ether of acetone 

 cyanhydrin. 



Hydrolytic Products of the Cassava Glucoside. 

 A portion of the purified extract, prepared as already described, was dis- 

 solved in water, a few cubic centimetres of 10-per-cent. hydrochloric acid added, 

 and the mixture distilled almost to dryness. The distillate gave the Prussian 

 blue reaction readily. To the remainder of the distillate, freshly-prepared 

 lead hydroxide was added, and the mixture allowed to stand for some time, 

 so that the hydrocyanic acid might be removed as lead cyanide. The filtrate 

 from this was redistilled, and the first few cubic centimetres collected. To 

 this were added a few drops of benzaldehyde, a similar small quantity of 

 potassium hydroxide solution, and enough alcohol to dissolve the benzaldehyde 

 added. On standing, the liquid deposited the characteristic crystals of 

 dibenzylideneacetone (melting point 112°). The volatile hydrolytic products 

 of the glucoside of cassava are, therefore, acetone and hydrocyanic acid, 

 identical with those of phaseolunatin, affording further proof of the identity 

 of the cassava glucoside with phaseolunatin. 



The Enzyme of Cassava Boot. 

 The enzyme was prepared in the usual way by extracting the ground dried 

 rind of bitter cassava root with water previously saturated with chloroform. 



LXXVIII. — B. N 



