Ernest Warren 
315 
and homozygous dominants we may approximately divide the individuals into the 
two groups by regarding the darker half as homozygous and the lighter half as 
heterozygous dominant. In this way we can form a correlation table between 
the parents and the heterozygous or the homozygous offspring. 
Using the symbols introduced by Prof. Bateson, namely RR homozygous 
recessive, DR heterozygous dominant, DD homozygous dominant, we have, on 
crossing individuals of the different gametic constitutions, the following theoretical 
resvdts : 
RR X DR = 1 DR (light purple) + 1 RR (white), 
DR X DR=1 DD (dark purple) + 2 DR (light purple) + RR (white), 
DRx DD=1 DD (dark purple) + 1 DR (light purple). 
In calculating the correlation it is advisable to use only parents and offspring 
of similar gametic constitution: thus, DR parents with DR offspring, or DD 
parents and DD offspring. RR parents and RR offspring being white are 
excluded. 
To accomplish this it is necessary in any given family, obtained by crossing 
together RR, DR and DD parents, to sort out the purple offspring into DR and 
DD individuals. To do this with complete certainty it would be necessary to self- 
fertilize each individual and obtain families. The labour involved is too great 
to be undertaken lightly. Owing to the fact that DR offspring are paler than 
DD offspring we may approximately divide the families into the U\o groups 
of individuals by taking the darker half (in the case of DR and DD parentage) 
or the darker one-third (in the case of DR and DR parentage) of the plants 
as being DD individuals in constitution and the remainder as being DR 
individuals. 
Thus for example: 
Parentage 
Family under Colour-Scale (see preceding table) 
VIII. C (1) X C (2) = DS X RR 
I. C(l) X C (1) = DR X DR 
IXa. C (1) X N (2) = DR X DD 
1, 2, 4 (DR), RR (excluded from table) 
1, 0, 1, 1, 2, 1 ( ?5 DD and f DR) = 1, 0, I DD + 1, 2, 1 DR 
1, 0, 3 ( h DD and ?, DR) = I, 0, 1 DD + 2 DR 
In this way a correlation table can be made with fair accuracy between DR 
parent and DR offspring. The paucity of the material, however, prevents much 
reliance being placed on the actual figures obtained. At most the figures can give 
only a rough indication of the strength of inheritance. 
When a heterozygous purple plant is crossed with a white plant (recessive) 
both purple and white offspring are produced. The coloured offspring, however, 
tend to be paler than the coloured parent. Thus the colour of the coloured 
