182 KARL B. KRISTOFFERSON 
Harrovist holds that this »ghoast-colour» depends upon the pre- 
sence in the white type of a factor for chromogen which is not able 
tc develop colour in the young flowers because of the absence of 
oxydase in the cell sap. Protoplasmatic oxydase will enter into the 
vacuole when the flower becomes older, and in co-operation with 
the vacuolar chromogen colour is now developed. It is true that in 
Aquilegia no fundamental colour factor has been found as is the case 
in Lupinus but, nevertheless, it is quite possible that the »ghoast- 
colour» depends upon a similar cause. The white flowering Aquilegia 
develops anthocyan in its vegetative parts as a rule. 
PLEIOTROPISM OF THE FLOWER COLOUR FACTORS. 
The anthocyan colour in the vegetative parts of the plants shows 
an interesting correspondence to the flower colour. Thus, the dark- 
blue and red flowering plants develop anthocyan on the upperside 
and the underside of the leaves, whereas the light-blue and the white 
types have the anthocyan located on the underside of the leaves. 
Both types have also anthocyan in the stem and in the leaf stalk. 
The intensity of the red anthocyan colour is also rather different. It 
is much darker in the dark-blue and the red forms. In summer it is 
almost impossible to discover any red colour in the leaves of the light- 
blue and the white types, while it is very easy to see in the dark- 
blue and red forms. Any difference in the development of anthocyan 
in the light-blue and the white plants is not to be seen. Consequently, 
the stronger or weaker development of the anthocyan is due to the 
red factor R. The light-blue factor B does not exercise any influence 
at all. Thus the white flowering Aquilegia is not even to the same 
degree as the white flowering Lupinus a »pure albino-form», although 
the latter is the case with other white types. 
When examining a Ds-line, that segregates in red and white 
flower colours, it is at once seen that the white flowering plants are 
much lower in stature than the red plants. 
In order to set down the correspondence between the height of 
the stems and the colour factors I have measured and calculated the 
segregating Ds-lines. It is possible that the segregation of the stem- 
height is due to special stem-height factors to a certain degree. If 
the number of true breeding D3-lines is too small, as is the case here, 
these factors might have so strong an influence on the means of the 
stem-height that the pleiotropical effect of the flower colour factors 
ie 
