THE OXIDASES IN RELATION TO PIGMENTATION 259 



Turning now to the question of what causes the decolora- 

 tion of the petals in the first instance, Wheldale and 

 Bassett have brought forward an ingenious analogy between 

 this and the decolorization of phenolphthalein solution, 

 rendered red by ammonia, which takes place in strong 

 alcohol, but is reversed on diluting with water. In this 

 case also evaporation to dryness leaves a coloured residue 

 Thus it seems reasonable to consider the observations with 

 anthocyanin as explicable in a similar manner namely, by 

 ionization changes, resulting in the production of a colour- 

 less undissociated compound which gives coloured ions. 

 There is, however, no direct proof of the correctness of this 

 view. 



It is nevertheless indisputable that plant tissues do con- 

 tain reducing agents. These are capable of causing the 

 disappearance of the colour of anthocyanin, as shown by 

 Keeble and Armstrong, and of both anthocyanin and the 

 blue oxidation product of guaiacum, as found by the 

 author (1913) for the leaf sap of Iris. In addition, the 

 above-mentioned collaborators (1913, 3) have drawn atten- 

 tion to decolorization of anthocyanin by nascent hydrogen. 

 It appears that such naturally occurring reducing sub- 

 stances must be responsible for some, at least, of the 

 destruction of anthocyanin which takes place in strong 

 alcohol. If it is thus easily reduced, it ought to be pos- 

 sible to oxidize it back again to the coloured condition, as 

 has been done by Bartlett with the pigment of Dioscorea 

 sp., though, so far as the writer is aware, there is no 

 proof that coloured products so obtained are identical with 

 anthocyanin. Undoubtedly the whole question has now 

 been settled once for all by Willstatter, who tried the effect 

 of alcohol upon pure crystallized anthocyanin, in which case 

 reducing agents were of necessity absent. Furthermore, 

 since it has been shown quite conclusively, as the writer 



