102 
S. IKENO : 
Table III. 
F 2 -generation (1917). 1 
F, -parent. 
No. of F 3 -offspring. 
Eed 
Orange 
Totals 
Eed x orange 
21 
6 
27 
Orange x red 
27 
11 
38 
Totals 
48 
17 
65 
Expected 
48-75±3-5 
16-25+3-5 
65 
F a - generation (1918). 
Colour of F 2 -parent 
No. of 
selfed 
plants 
Eed. 
6 
Eed. 
5 
Orange . 
6 
* Orange f. 
1 
* Orange . 
1 
No. of F., -offs pring. 
Eed 
Orange 
Totals. 
f 
35 
0 
35 
1 Expected 
35 
0 
35 
J 
85 
23 
108 
I Expected 
81+4-5 
27±4-5 
108 
0 
70 
70 
I Expected 
0 
70 
70 
1 
1 
3 
4 
( Expected 
0 
on 
4 
4 
op; 
I Expected 
18-75+2 2 
6 25+2 2 
25 
liomo : 7‘30 lietero) (s. Table III), so that the ratio of individuals in these 
two classes of reds does not well agree with what we might expect theoretical¬ 
ly, but much importance should not be laid on this fact in view of the small 
number of individuals. 
From the above experiments we see that the difference between orange 
and red is due to one factor which we call H and which changes the orange 
caused by C into red. 
1 Though, as may be seen from the results in F a -generation, there are two kinds of reds 
(homo- and heterozygous), they are collected in F> under the class red, because they are not 
exactly distinguishable from each other by their external appearance. 
+ The parent prefixed with an * denotes one which has produced some unexpected individuals 
among its offspring. Thus, for instance, in this No. we find besides 3 oranges 1 red which was 
quite unexpected. For the discussion of these phenomena s. pp. 121. 
