The Inheritance of Quantitative Characters in Maize 17 
n = l 1 2 1 = 4 
n = 2 1 4 6 4 1 = 16 
n = 3 1 6 15 20 15 6 1 = 64 
n = 4: 1 8 28 56 70 56 28 8 1. . = 256 
n =5 1 10 45 120 210 252 210 120 45 10 1 = 1024 
Let us now note a few of the practical difficulties in inter- 
preting breeding results that may follow this method of 
inheritance. In the theoretical example that we have used for 
the sake of clearness, it was assumed that there were no non 
heritable fluctuations due to environment. Unfortunately this 
is not the case in nature. Fluctuations are everywhere present. 
They would obscure the classes to which individuals belong 
even if these class differences were very large. And since 
they are usually small, the change of individual form due to 
environmental causes makes it impossible to separate an F 2 popu- 
lation into the true classes to which each belongs theoretically. 
Nor is this the whole trouble. If the table showing the expected 
results Avith two pairs of size characters is examined, it is found 
that not all the individuals that belong to a particular size class 
have the same zygotic formula. For this reason one cannot pick 
out zygotes of a certain size and expect them to breed the same. 
Their potentialities are likely to be different. Furthermore, 
practical breeding results are undoubtedly complicated by cases 
of correlation. This correlation need not be gametic, tho such 
cases in all likelihood do occur; they may be merely physiological. 
For example, a maize plant might have the gametic possibilities 
of small plant size and large ear size, but it would be foolish to 
expect that a plant capable of only a limited amount of develop 
ment could bear as large an ear as if it were, as a whole, capable 
of greater size development. Thus it must not be expected that 
theoretical possibilities are always expressed perfectly in nature, 
any more than it should be expected that theoretical physical 
calculations concerning known laws should agree perfectly with 
experimental data. The reproductive cells of plants and animals 
may indeed be described as mosaics of independently trans- 
missible factors, but a plant or animal certainly can not be 
considered a mosaic of independent unit characters. These 
far- tors that appear to be independent in heredity act and react 
upon one another in complex ways during their development. 
There is a means, however, by which it can be determined 
whether segregation does or does not occur in cases of seemingly 
blended inheritance. This is by comparing the variability* of the 
* In order to understand the data from these experiments it is neces- 
sary to be familiar with the meaning of several mathematical terms used 
