The Inheritance of Quantitative Characters in Make 31 
which can vary in any one of eight spikes will show a greater 
degree of fluctuation than one which can vary only in any one 
of four spikes. For this reason it is likely that strains with a 
high number of rows will never show the low variability seen in 
strains with a low number of rows. 
Taking into consideration the probable amount of fluctuating 
variation, several interesting points are presented in the tables 
showing the frequency distributions of the number of rows per 
ear in the various crosses. 
Tables 3, 4, and 5 show the results from three different crosses 
in which No. 5,* an 8-rowed white flint, was crossed with other 
varieties. In Table 3 the male parent was No. 6, the well known 
Learning dent. In the particular strain from which this male 
parent was taken, the modal condition (that is, the class of 
greatest frequency) was sixteen. Four F 2 families were grown, 
of which only one was markedly more variable than the F 1 gen 
eration The modal condition of FJ12) was intermediate be- 
tween the two parents. In three of the F 2 families, the modal 
condition was ten; in the remaining family it was twelve. In 
Table 4. the male parent, a white dent, was also typically twelve 
rowed. The F x and F 2 generations were comparable to those 
in Table 3. While the F 2 generation was much more variable 
than the ¥ 1 generation, no great confidence can be placed in the 
results, owing to the meager numbers of the F t generation. 
Table 5 shows a peculiar result. Here No. 5 was the male 
parent, while the female parent was a 12-rowed flint which is 
very low in variability. Both the Fj and F 2 generations were 
typically twelve rowed. 
These data form the basis for several speculations. In two 
crosses the typical condition of the hybrid was intermediate. 
In this last cross the 12-rowed condition appears to be domi- 
nant. Is this due to the fact that the 12-rowed parent was the 
mother? Is it due to the fact that the maternal individual was 
gametically 16 rowed? Is it due to there being few differences 
between the parents by which complications thru correlation 
might arise, for both are flints? It seems probable that the last 
two suggestions are likely to be at the basis of the phenomenon, 
but there is no way of deciding from the evidence at hand. 
Table 0 merely shows that this same 12-rowed variety when 
crossed with another variety which is typically twelve rowed 
shows no greater variability than one of the parents. 
Table 7 gives the data from a cross between an 8-rowed 
flint and a very variable 12-rowed dent. The description of the 
Fj generation was not taken at the time, as the cross was made 
primarily to study another question. The omission is partially 
* For description of these varieties see page 29. 
