212 ANTHOCYANINS AND GENETICS 



for there may be more than one flavone. In fact it is impossible to 

 know until the flavones are isolated and analysed as in Antirrhinum, 

 what the chromogen is. 



As regards the colour-producing factor, which we have called R, 

 and which is carried by the white in Antirrhinum and by one of the whites 

 in Lathyrus, it has been suggested that it is an oxidase or oxidising 

 enzyme which acts upon the chromogen with the formation of pigment. 

 The evidence in favour of this hypothesis is not very convincing. If 

 the hypothesis be true, then the ivory variety in Antirrhinum, one of 

 the whites of Lathyrus and Matthiola, and possibly the recessive whites 

 of Primula should be without oxidase. This is not yet confirmed, for 

 all the varieties which have been tested contain oxidase. Nevertheless, 

 this statement needs amplification. The R factor, it is true, produces 

 colour, but there are- other factors to be considered later which 

 modify this colour, making it deeper or paler. These may also be 

 oxidising enzymes, so that it is essential, when testing for R,, that one 

 should use carefullv selected individuals. In Antirrhinum, for instance, 



t/ * 



the individual should be yyiiLLttddbb, in Lathyrus, ccRRbbdededidi, 

 and so forth ; individuals for the finding of which several years' breeding 

 experiments are often needed. In the testing of Lathyrus, Matthiola, 

 Antirrhinum and Primula for oxidases, adequate precautions to secure 

 the right individuals on these lines have not been taken; the same 

 criticism holds good for the testing for chromogens in Primula, Lathyrus 

 and Malthiola. In addition, even if oxidases prove to be connected 

 with colour, it seems likely that there may be in the plant other oxidases 

 controlling reactions essential to metabolism, in fact it is difficult to 

 conceive of a plant albino or otherwise : which is entirely free from 

 oxidases. Until some reagents can be found which differentiate between 

 oxidases, it is impossible to draw safe conclusions. 



Nevertheless there appears to be some connection between oxidase 

 distribution and colour distribution. For Keeble & Armstrong 

 found that in Primula flowers where colour is inhibited, the oxidase 

 reaction is inhibited too, but could be detected after removing the 

 colour inhibitor with certain reagents. The author (213), also, found 

 that the whites extracted from differently coloured varieties of Matthiola 

 and Lathyrus, though all giving the oxidase reactions, did not give 

 the reaction to the same extent, that is, it varied in intensity. 



. The foregoing suggestions are based upon the assumption that 

 anthocyanin is formed from a flavone by oxidation. Willstatter, as 

 we have seen in earlier chapters, considers the process of colour formation 



