The Production of Anthocyanins and Anthocyanidins. 445 



could any trace of anthocyanidin be found in any of the plants examined,* 

 there remains but one way of explaining their absence if Miss Wheldale's 

 hypothesis is to be retained ; namely, by assuming that the rate of forma- 

 tion of anthocyanin (glucoside) from anthocyanidin and sugar is greater 

 than that of the production of anthocyanidin from chromogen by oxidation, 

 and that these reactions take place under similar conditions. 



If this were correct one would expect that by taking the yellow glucoside, 

 hydrolysing, then reducing without removal of sugars, the anthocyanidin 

 produced would combine with the sugar present to form an anthocyanin. 

 This is not the case. Evidence all tends to show that Miss Wheldale's 

 view can no longer be accepted as explaining the available facts. 



The reaction found so useful in determining the presence or absence of 

 glucosidal or non-glucosidal anthocyan, and already described by Willstatter 

 and Everest (loc. cit.), yields a very ready means of distinction between 

 these two classes of compounds, and has led to important results in the 

 present investigation. It depends upon the facts that anthocyanidins (non- 

 glucosides) are taken quantitatively from aqueous acid solutions, preferably 

 sulphuric acid, by shaking with amyl alcohol, whereas anthocyanins 

 (glucosides) remain quantitatively in the aqueous acid when similarly 

 treated ; and, further, that the non-glucoside in amyl alcohol, shaken with 

 sodium acetate solution, remains quantitatively in the amyl alcohol, but on 

 shaking with sodium carbonate solution it is quantitatively carried down 

 into the aqueous layer. 



The author has been able to show the production of anthocyanins from 

 yellow glucosides, and that in the cases where hydrolysed solutions were 

 taken only anthocyanidins were produced. That no intermediate formation 

 of anthocyanidins occurred when anthocyanins were obtained was shown 

 by carrying out the formation under amyl alcohol. Where glucoside 

 yellow pigments were used the anthocyanin appeared as usual, but no 

 anthocyanidin passed into the amyl alcohol ; when a hydrolysed solution 

 was similarly treated the amyl alcohol rapidly took up all the anthocyanidin 

 as it was produced. 



This makes it necessary to abandon the assumption suggested above as the 

 only explanation available if Miss Wheldale's hypothesis is to remain. 



A number of papers have been published upon this subject by Keeble, 

 Armstrong, and J ones, f and they conclude that the anthocyan pigments are 



* Willstatter and Everest, ' Annalen,' vol. 401, p. 189 (1913). 



t 'Boy. Soc. Proc.,' B, vol. 85, p. 214 (1912); B, vol. 86, pp. 308 and 318 (1913) ; 

 B, vol. 87, p. 113 (1913) ; and Keeble and Armstrong, ' Journ. Genetics,' vol. 2, part 3, 

 p. 277 (1913). 



