STUDIES ON MENDELIAN FACTORS IN AQUILEGIA VULGARIS 179 
have this statement from gardeners, in any case, though I myself 
have made no attempt in this direction. 
My experiments begun in 1915. This year several plants of 
Aquilegia vulgaris were isolated in the Botanical Garden of Lund. 
Some crosses were also made. Some of the isolated plants bred true 
in later generations, while other showed segregation; it is especially 
upon one of these latter ones I have worked further. Seeds from 
the parchment bag isolations and from the crossings were sowed the 
following year but, the locality being unsuitable, only few plants 
flowered in 1917, and therefore the examination of the material was 
put off till 1918. 
THE INHERITANCE OF THE FLOWER COLOUR. 
When the D.-generation* of one plant with self-coloured dark- 
blue flowers was examined it was found that the flower colour showed 
segregation in 22 dark-blue, 9 red (red violet), 7 light-blue and 2 white. 
It is at once seen that this segregation is a dihybrid one due to two 
factors. The factor B gives the light-blue colour, and the factor R 
the red. B together with R gives the dark-blue colour, and when 
both factors are absent white colour results. The genetical formulae 
of the original plant was BbRr and the segregation in D2 must 
be : dark-blue, red, light-blue and white in the ratio 9:3:3:1. The 
ratio observed 8,8 : 3,6 : 2,8 : 0,8 pro 16 agreed very well with the 
thecretical one, as shown in table 1. 
Some of the D2-plants bred true in Ds; other: showed segregation. 
Table 2 shows the distribution of the true-breeding and the segregating 
D3-lines. The observed and the theoretical values agree very well. 
The tables 3—7 show the segregation of the colour in the segre- 
gating D3-lines. The correspondence between the theoretical and the 
observed values are also here rather good; in all cases the deviation 
is smaller than its standard error. 
This colour segregation is rather unique. Pure blue flower colour 
together with violet and light-blue do not occur, as a rule. When 
a variety is blue it is generally found to be a recessive and not the 
type, which Harrovıst (1921) also remarks in his paper on Lupinus. 
So far as I know HarnLovısr and VESTERGAARD (1919) are the only 
investigators who have found a synthesis of blue colour from crossings 
1 Following HERIBERT-NILSSON (1920) the progeny. of one single plants is 
here called D1, D2 etc.; Pi, P2, F1, F2 etc. are reserved for crossings. 
