308 Miss M. Wheldale and Mr. H. LI. Bassett. 



well as any other acid, which would not be the case if an enzyme were 

 responsible for the restoration of colour. 



In any of these cases when the anthocyanin restored by acid is made 

 alkaline, the greenish colour reaction of anthocyanin is obtained, showing 

 that the restored colour is actually due to anthocyanin. The greenish 

 reaction is also produced directly when a drop of a solution of caustic soda 

 in absolute alcohol is added to the alcohol containing the decolorised petals. 



If water is boiled to expel oxygen and carbon dioxide, and, while still hot, 

 a stream of hydrogen is bubbled through it, this water, while the hydrogen 

 is still passing, restores the colour to decolorised petals. In this case the 

 medium is neutral. 



It is not conceivable that oxidation can take place in all these experiments, 

 particularly in that with dry hydriodic acid gas. Clearly also water is not 

 necessary for the change, and another explanation for the restoration of 

 colour must be sought. 



Further, if reduction is the cause of decolorisation, the conditions in some 

 of these experiments are exactly those most suited for the continued stable 

 existence of the leuco-compounds, so that it would seem that this explanation 

 must also be abandoned. 



There may be a reducing agent present in the petals, but its reducing 

 power cannot be responsible for the loss of colour in alcohol. 



In support of their theory that reduction is the cause, Keeble, Armstrong, 

 and Jones, in a later paper* quote the fact that an extract from the petals is 

 reduced to a colourless state by treatment with zinc dust and acid, and that 

 the colour is restored by exposure to air. We would note in passing that 

 this does not seem to be simply a reducing action, as we find that the 

 restored colour is much fainter with acetic than with sulphuric acid. This 

 observation has been made previously by Kastle,f who also does not consider 

 it simply a reducing action. Untreated anthocyanin gives exactly the same 

 colour with acetic as with sulphuric acid. 



A point we wish to emphasize, however, is that we find the slightly acid 

 solution to be easily decolorised by warming with a little hydrogen peroxide 

 and colloidal platinum. The colourless oxidation product so formed is 

 unstable, and the colour is restored if the solution is made more strongly 

 acid. 



* Keeble, F., Armstrong, E. F., and Jones, W. N., " The Formation of Anthocyan 

 Pigments in Plants.— Part VI," ' Eoy. Soc. Proc.,' 1913, B, vol. 87, p. 113. 



t Kastle, J. H., " A Method for the Determination of the Affinities of Acids Colori- 

 metrically by Means of certain Vegetable Colouring Matters," 'Amer. Chem. Journ.,' 

 1905, vol. 33, p. 46. 



