46 Nebraska Agricultural Exp. Station, Research Bui. 7. 
first five internodes of the F2 plants, 26.36 ±0.54 mm., is very 
nearly exactly intermediate between the means of the parents, 
30.09 ±0.77 and 22.65 ±0.32 for July and Snowflake, respectively. 
But both the standard deviation and coefficient of variation for 
Fo are less than these constants for one of the parent races, July, 
tho they are considerably greater than those for the other parent 
race, Snowflake. Of itself, the fact that the coefl^icient of varia- 
tion for F2 is slightly greater than the average of the coefficients 
for the parent races cannot be used to show a segregation in F? 
of genetic factors for internode length, unless, by the same reason- 
ing, we are forced to admit also that there is segregation of an 
even greater number of genetic factors in case of the parent race 
July. While it is not unlikely that July is still heterozygous for 
factors concerned in internode length, it is highly improbable 
that it is heterozygous for more factors than Fi plants of a cross 
between it and the very distinct race, Snowflake. If the high 
coefficient of variation for Fo, 25.61 ±1.55 per cent, is due to 
segregation of size factors as seems probable from other consider- 
ations, the still higher coefficient for July, 32.67 ±2.64 per cent, 
is yet to be explained. 
It seems likely that the great variation exhibited by July, 
with respect to mean length of the first five internodes, is a chance 
relation of its peculiarities of growth in the early stages of its 
development to the somewhat uneven soil moisture conditions 
of the different parts of the garden as noted above. From the 
data presented in the discussion of methods of determining inter- 
node length, it will be recalled that, tho July is extremely vigorous 
in growth after once growth is well started, its first two or three 
internodes are unusually short — shorter even than those of the 
much less vigorously growing Snowflake. The mean lengths of 
each of the first seven internodes as determined from measure- 
ments of 83 plants of July and 78 plants of Snowflake are repeated 
here: 
Since the first internodes of July are so very short and some- 
what later ones so very long, the change in internode length is 
necessarily extremely rapid. If now this abrupt acceleration in 
growth-rate should begin in one plant with the fourth internode 
and in another with the sixth — as might easily happen if, owing 
to a slight difference in soil moisture, the one plant germinated a 
little later than the other, so that the one had developed only 
three internodes while the other had developed five at the time of 
a heavy rainfall — the mean length of the first five internodes 
Internode No 
July 
Snowflake 
1 2 3 4 5 6 7 
19 14 23 41 56 78 115 
26 17 20 24 25 28 38 
