CORN BREEDING 23 
recessive. Thus, for example, the recessive dwarf condition is in- 
ferior to the normal. Similarly, albino seedlings can not survive 
at all, whereas the condition controlled by the dominant factor 
allelomorphic to that for albinism produces normal plants, other 
things being equal. Nearly all of the known factors that determine 
a condition in corn strikingly unfavorable to growth are recessive 
to their respective allelomorphs. The same condition also obtains 
in large measure in other cross-fertilized plants and in animals. 
It is easy to see why this should be so. Assume a homozygous 
strain of normal corn as a starting point. Such a strain would 
breed true because the two members of every allelomorphic pair 
would be alike. If now a mutation were to occur in some factor, 
the new factor (mutant) would have to be either more or less favor- 
able for growth or indifferent in its effect. It also would have to 
be either dominant or recessive to its allelomorph, or dominance 
could be lacking. 
If the new factor were dominant the character controlled by it 
would come into immediate expression and, if more favorable to 
growth, would tend to be perpetuated through the forces of natural 
or artificial selection. If, however, such a dominant mutant were 
less favorable for growth, the tendency would be to eliminate it 
at once. Thus, in an extreme case, if albinism were to originate 
as a dominant mutation, every plant carrying this new factor would 
be an albino which would die in the seedling stage and albinism 
would be eliminated from the strain. 
Assume now that the mutant were recessive. It would not be 
expressed at all. in the first generation. When it did come into 
expression, if it were more favorable for growth than the original 
factor, the tendency would be to eliminate the older factor. On 
the other hand, such a recessive mutant would be carried along in 
the heterozygous plants in which the character it controlled would 
not be expressed even if less favorable to growth, although there 
would be a tendency to eliminate it. 
With intermediacy or partial dominance the tendencies would be 
the same in each case, but less marked. If there were little or no 
difference between the two factors in their effect on growth there* 
would be no selective action, and both would tend to remain in the 
variety. 
In the centuries during which corn has grown there has been 
time for innumerable mutations to occur and for selective action to 
eliminate such as were dominant and unfavorable to growth. Due 
to its extensive cross-fertilization, recessive factors could be carried 
along in the heterozygous condition, although when unfavorable to 
growth their proportion in the population as a whole would be 
kept at a minimum. Our present varieties of corn consequently con- 
tain an assortment of factors, some of which are more favorable 
and some of which are less favorable to growth. The favorable 
factors generally are dominant and in excess of the less favorable 
ones, which in general are recessive. 
Such unfavorable recessives come into expression only in plants from 
kernels resulting from the fusion of an egg and a sperm both of 
which carried the specific factor. In ordinary crossbred corn this 
does not occur to any great extent for any one character. The total 
