116 Chemist ri/ of Mendelian Factors for Floicer -Colour 



Pure anthocyanin was prepared from five varieties of Antirrhinum 

 in the following way (Wheldale and Bassett, 35). Crude pigment was 

 obtained by the method described for yellow and iv(jry varieties. Since 

 anthocyanin only occurs in the epidermis, while apigenin is found in 

 the inner tissues of the corolla, all crude pigment consisted of mixtures 

 of anthocyanin and apigenin, and, in some cases, of luteolin in addition. 

 Anthocyanin was purified from flavones by extracting the mixture for 

 several months with boiling ether. The residue of crude anthocyanin 

 was then dissolved in the mininuim amount of alcohol and precipitated 

 by ether in which it is insoluble. The dried precipitate was again 

 extracted with ether to remove all traces of flavone. In this way 

 anthocyanin was obtained in a purer condition than by crystallisation, 

 since fi'om a mixture flavone and aiithoc\-anin readily crystallise out 

 together. 



Combustions of pure anthocyanin prepared from rose dore and from 

 bronze showed the anthocyanin pigments to be identical in both ca.ses, 

 the difference of colour in the flowers being merely due to the presence 

 of luteolin in the latter case. 



In the same way combustions of pure magenta anthocyanin from 

 magenta and from crimson showed that crimson in the flower is a 

 mixture of magenta anthocyanin and luteolin. The same magenta 

 anthocyanin was also obtained from a variety — ivory tinged with 

 magenta — which is formed from magenta by loss of a deepening factor. 

 Hence the anthocyanin pigments in the tinged and full-coloured 

 varieties are identical. 



From analyses of anthocyanin from both red and magenta varieties, 

 both pigments were found to contain much higher percentages of oxygen 

 than the flavone : 



C 



Red anthocyanin ... .'5r81°/„ 

 Magenta anthocyanin .50-.50 

 Apigenin ... ... 66'66 



A determination of the molecular weights of the two anthocyanins 

 showed them to be of the order of magnitude of 700 for magenta and 

 570 for the red. Hence, if derived from flavones, of which the molecular 

 weight is about 270, changes other than oxidation must be involved. 

 It is most probable that there is also condensation of two or more 

 molecules of flavone or condensation of a flavone molecule with other 

 aromatic substances. If the latter be the case, condensation with one 



