THE OXIDASES IN RELATION TO PIGMENTATION 269 



The molecular weights of the red and magenta antho- 

 cyanins were found to be 572 and 717 respectively by means 

 of ebullioscopic determinations in ethyl alcohol. Since the 

 results of the combustions may most simply be expressed 

 by the formulae, C 8 H 9 O 5 for the red, and C 15 H 18 O 10 for the 

 magenta, the boiling-point determinations indicate their 

 formulae to be respectively 3 (C 8 H 9 O 5 ) i.e., C 24 H 27 O 15 , 

 mol. wt. 555 and 2 (C 15 H 18 O 10 ) i.e., C 30 H 36 20 , mol. 

 wt. 716. Evidence derived from analyses of the lead 

 salts supported these values. It was also ascertained 

 that the red contained twelve hydroxyl groups, and 

 the magenta anthocyanin fifteen. These figures and the 

 high values for the percentage of oxygen in anthocyanins 

 led Wheldale to conclude that the production of the latter 

 in the plant was due both to condensation and oxidation 

 of the flavones. 



Assuming such flavones to be the parent substances of 

 the anthocyanins of Antirrhinum and on this point the 

 chemical and genetic evidence is decisive it may be re- 

 garded as well established that the anthocyanins examined 

 by Wheldale are more complex than those studied by 

 Willstatter, and are richer in oxygen. They cannot, there- 

 fore, be reduction products, though other anthocyanins 

 may quite conceivably be so. 



MENDELIAN EXPERIMENTS BEARING ON ANTHOCYANIN 

 FORMATION. 



These researches illustrate the fruitfulness of the com- 

 bination of chemical methods with those of Mendelian 

 analysis. By the latter it had been shown that the 

 varieties of colour occurring in Antirrhinum majus necessi- 

 tated for their elucidation the grouping of at least six 

 factors. It remained for chemical research to show that 



