246 ON THE INHERITANCE OF THE FLOWERING TIME IN PEAS AND RICE 
-Crosses with the kite flowering pure line of G- P. 
VIII X 12.1 
VIII X 12.4 
12 X VIII.i 
12 X VIII. 2 
Sprouting began on the 17th and ended on the 19th of the month. There 
were a few which sprouted after the 20th, but the flowering dates of such 
delayed individuals were not recorded. As already mentioned, from this 
year the experiments were carried on in the nursery ground. 
In Tables 9 and 10, Pis. XII — XIV, we show the frequency distribution 
of the F3 raisings of the year 19 12, arranging all families according to the 
similarity to their variation types. The white dots indicate the white-flower- 
ed individuals and the black dots the red-flowered ones. 
In 1913. 
In 191 3, the F3 offsprings of the families XVI x 3.2; XVI x 3.3 and 
3 X XVI. 2 were raised. The seeds were sown on May 9, and sprouting took 
place during the 3 days from the 21st to the 23rd of the month. 
The results of this raising are shown in Table 1 1, Pis. XV — XVI. 
Chief Facts from F3 Raisings. 
In the P'g raising of 19 10 (Table 3) where only the progenies of those 
individuals were raised which flowered within the variation ranges of the 
parent varieties ([^age 236), we have seen that there were several types of varia- 
tion of would-be constant (homozygous) and variable (heterozygous). Now 
in the present V.^ raisings, where the progenies of almost all individuals of 
each of the selected F"^ families were grown, we see greatly complicated 
variation types, and can here again distinguish two types of variation, the 
constant and variable. 
At first, we shall treat of the constant families. In Table 9 (Pis. XII 
and XIII), 19 families, from No. I to No. 19, are what might be taken as con- 
stant. Judging from the variation modes and means of these constant families, 
Nos. I, 2 & 3 seem to be the car/}' constant families, having the characteristics 
