FACTORS DETERMINING COLOUR 253 



may be made of certain facts in order to illustrate the inter- 

 relationship between this branch of botany and chemistry. 



In many plants the colour of the flowers depends on 

 various factors : 



C, a chromogen which is not necessarily coloured, and 

 which is, in all probability, a glucosidal flavone. 



E, an oxidative enzyme which acts upon C to produce a 

 red colour. 



e, another enzyme which acts upon the red pigment, due 

 to the action of E, and further changes it to another pigment 

 so that a different colour results. 



A, an antioxidase which inhibits the action of E. 



R, a reductase which neutralizes, as it were, the action of E. 



If a flower, say, of Latkyrus, only possesses C or E, then 

 the colour will be white or pale yellow, according to the colour 

 of the chromogen, if present. If the flower with the factor 

 C be crossed with a flower with the factor E, then the colour 

 of the flowers of the offspring will be red, or a deeper colour 

 if e also be present. If either A or R be present, then there 

 will be no difference in the flowers of the offspring as com- 

 pared with the parents. 



By microchemical methods it is possible to ascertain the 

 nature of many of these factors present in the plant, and by 

 such means Keeble and Armstrong* have established many 

 facts relating to the distribution of oxidases in Primula 

 sinensis and their relation to the formation of pigments. 



The method employed by these authors is the treatment 

 of the tissue with suitable reagents dissolved in media, termed 

 hormones, which solvents cause the plasmatic membranes to 

 become permeable to the reagents, and also renders active 

 the enzymes contained within the cell. One such substance 

 is alcohol. 



To take an example : <z-naphthol is commonly used as an 

 indicator for oxidases ; treatment of the petals of Primula 

 sinensis with an alcoholic solution of a-naphthol decolorizes 

 the tissues, but on then treating the tissue with hydrogen 

 peroxide a coloration obtains, and a comparison of this pre- 

 paration with the original indicates that the distribution of 



* Keeble and Armstrong : " Proc. Roy. Soc., Lond.," B., 1912, 85, 214, 460. 



