206 THE INHERITANCE OF CHARACTERS IN RICE, II 
fig. 6). In F,a /2: 3: 1 ratio of purple: red: white is obtained as shown 
by the following figures :— 
Purple ted White 
PR-+-Pr p R pr 
22 tamilies ss Ai 4,540 ppl Na) 449 
Calenlated 12:3: 1 oe 400 Sean el LOSE ae OKS 
The purple group is a very mixed lot forthe same variation exists as 
described for F, of the purple white cross and, in addition, various degrees of 
purple-red appear. Some of the latter, where the purple is reduced to the 
greatest extent, nearly approach ordinary red though they are separable in 
well-developed grain. 
Purple F, plants give four types of family in F.—pure purples, 3 : 1 ratios 
ot purple: white, 3:1 ratios of purple: red, and the same as F, again. Some 
of the pure purples show considerable reduction of the purple colouring, as 
noted above in the purple < white lots, and in such cases the P P R R type 
appears as purple-red like the F, Pp Rr. 
It may be noted here that the factor P falls into the same linkage group 
as L, G, S and A, the factors responsible for purple colouring in the 
internode, glumes, stigma and axil, as described in Part I, pp. 91-97. 
Several other factors have also been found to fall into the same group and 
their relations will form the subject of a future paper, 
Pui ple and brown rice. 
Just as the pigmentation factor .f is necessary in the production of red 
rice so also it 1s necessary in the production of purple rice. In its absence the 
factor P produces brown rice (Plate V, fig. 7), corresponding to the grey-brown 
rice of the factor R (fig. 8). 
This behaviour has not been worked out in detail, as in the case of grey- 
brown rice, but is sufficiently evident from the results of a natural cross that 
proved to be of the constitution Pp Rr Aa Nn. F, gavea 9:7 ratio of 
pigmented : unpigmented. The former group comprised purple, red and white 
and the same three types, together with brown and qrey-brown, occurred in the 
unpigmented group. 
Accurate counts were not made as there was some difficulty in separating 
the rice groups with certainty. The same difficulty was experienced in the 
more complicated familiesin F;. The degree of ripeness affects the rice colour 
considerably; also, apparently, the rapidity with which maturation takes place 
has some effect on the degree of development of the colour, For this reason 
