YUZO IIOSHINO 
type of the progeny of aabB should be that of the progeny of a monohybrid, 
varying between the early constant (aabb) and the early intermediate constant 
(aaBB)j \\'hile that of the progeny of aAbb, being also of a monohybrid type, 
should vary between the early constant (aabb) and the late intermediate con- 
stant (AAbb)- There should not occur any individuals which flower in the 
same period as the late constant. In actual F.^ raisings, we have seen that 
none of the variable families descended from the early flowered plants 
produced such a wide variation range as to reach the range of G- P- (page 
248). It is quite difficult to divide all variable families from the early flower- 
ed F.^ plants into two distinct categories according to their variation types, as 
the actual variation types shown in Tables 9, 10 and i i are too complicated 
and irregular. But we may say, that this difficulty in distinguishing two 
types of variation does not positively conflict with the proposed hypothesis, 
because we are concerned with experiments on such a physiological character 
as the flowering time, which even in homozygous progenies varies some- 
what, according to environmental influences. 
Among I".j variable families descended from the late flowered heterozy- 
gous F^ [plants, we should expect three different types of variation, that of 
the progeny of aAbB, which varies from the early constant to the late con- 
stant, as in F^ family; that of aABB, which, being of a monohybrid type, 
varies between the early intermediate constant and the late constant ; and that 
of AÄ.bBj which is between the late intermediate constant and the late constant. 
In Tables 9, 10 and 1 i, it is difficult to divitle sharply all variable families 
descended from late flowered plants into three types, but it is certain that 
there are some families which represent each of the three expected types. 
The upper end of the variation ranges of all F^ variable families des- 
cended from the late flowered F.^ plants reaches the range of G- P- except a 
few cases as described in (2) and (4) on pages 248 and 249. Among these 
exceptional families, those described in (2) we have regarded already as the 
l)rogenies of individuals which were in the early flowering group genetical- 
ly but fell into the late group, owing to some environmental influences, and as 
