86 



THE AMEBIC AN NATUEALIST [Vol. XLIV 



is hypostatic to all the red shades, ' ' although the magenta 

 colors are shown to be epistatic to red. This isolated 

 statement regarding the bine color in Primula Sinensis 

 is not supported by any data, and I do not know the 

 chemical relation between it and the magenta colors. 

 Miss Wheldale (1909), who discusses at some length the 

 color series in Primula material secured from Bateson 

 and Gregory, makes no mention of the occurrence of blue, 

 though she ascribes the production of magenta and 

 crimson to the action of a "bluing factor" upon red 

 anthocyan. 



Upon comparison of my bluish-purple Lychnis with 

 the colored plates of Primula given by Bateson (1909, 

 p. 294) I think the Lychnis color should be classed as a 

 light magenta rather than a blue, as there is a decided 

 reddish element in this Lychnis color. If Bateson 's iso- 

 lated statement that "blue is hypostatic to all the red 

 shades" in Primula is correct, then that color corre- 

 sponds in its behavior with this light magenta color in 

 Lychnis. I have not made a thorough investigation of 

 the chemical relations of the two types of purple in 

 Lychnis, but have demonstrated by a few preliminary 

 tests that the reddish-purple color is converted to bluish 

 when treated with alkalies, and that the light bluish- 

 purple is made as bright red as the red-purple type on 

 treating with weak acids, thus indicating a very simple 

 relation between these two colors. 



Although the relation between the two types of color 

 in Lychnis is just the reverse of that exhibited by prac- 

 tically all other plants in which similar colors have been 

 studied, I am led to essentially the same conclusions re- 

 garding the method of color determination, as those de- 

 rived from the extensive studies which have been made 

 on Lathyrus, Matthiola, etc. The production of the 

 "lowest stage" of color, i. e., the color which results from 

 the combined action of the least number of genes, i 

 to the interaction of two independent factors or £ 

 either of which produces no color when not associates 



