FLOWER AND SEED COLOUR IN LUPINUS 335 

table weather conditions during the early vegetative season kept the 
young plants small and poorly developed for a long time). The di- 
stinct development of the colour made it easy to notice the difference 
in F, of cross 7 between the whites in the families segregating in blue 
and white, and the whites in the families segregating bluish red and 
white. The former were more bluish than the latter. A classification 
of ‘the different white types coexisting in families of di-hybrid segrega- 
tion was then tried. 62 white individuals from such families were 
divided in two groups. In the first group individuals were put ha- 
ving the same colour as the one noted in the families segregating in 
blue and white; the second group includes whites coloured in the 
same shade as in families segregating in bluish red and white. The 
numbers of individuals in these groups were 41 : 21; the expected pro- 
portions, according to the ratio 3: 1, is 46,5 : 15,5. Thus it is clear that 
the factors B and V also determine the colour of the white types. 
The question now becomes of interest whether the phenotypical 
differences between the white types in Lupinus and other genera in- 
dicate a difference as well in the factorial constitution. It is possible 
that the modifying factors B and V have a colour effect of their own 
without the co-operation of the fundamental factor. It is also possible 
that the white type does not represent an albino form but a dilution 
form, brought into existence by the absence of a full colour factor. 
These hypotheses are probably both erroneous. The colouration in 
question certainly does not indicate a diverging factorial constitution; 
it depends rather on purely physiological peculiarities. The colora- 
tion of the white flower belongs probably to the same category of 
phenomena as the shading of the colour noticeable in the coloured 
types in Lupinus when old. In the genus Anchusa, where also the 
colour of the flowers changes with age, no coloration of the white 
type is seen. Thus there are marked differences between the white 
types of Lupinus and of Anchusa; it does not invalidate, however, the 
parallel drawn in the above with regard to the shading of the coloured 
and white types. It is probably due to physiological changes in the 
tissues of the plant. A change in the permeability of the vacuolar 
membrane suggests itself. It is conceivable that a substance entering 
the vacuole may be endowed with the power to change the anthocyan 
of the cell sap, thus compensating the colour factor, which has been 
found to be absent. If we, accepting WHELDALE’s hypothesis of colour 
formation, assume the presence in the white type of a factor for 
chromogen and the absence of the factor for the development of 
