ngth of{ F_1 | 1| 7| 3| 2| 1| | | | 2| | 1| | 1|
ulna { B.C. | 2| 4| 2| 2| | | 1| 1| | | | | |

He found that the variability was smaller in the first generation than in
the second generation (back cross). This is what is expected if several
factor-differences were involved, because the hybrids of the first
generation are expected to be more uniform in factorial composition than
are those in the second generation which are produced by recombination of
the factors introduced through their grandparents. Excellent illustrations
of the same kinds of results have been found in Indian corn. As shown in
figure 85 the length of the cob in F_1 is intermediate between the parent
types while in F_2 the range is wider and both of the original types are
recovered. East states that similar relations have been found for 18
characters in corn. Emerson has recently furnished further illustrations of
the same relations in the length of stalks in beans.

[Illustration: FIG. 85. Cross between two races of Indian corn, one with
short cobs and one with long cobs. The range of variability in F_1 is less
than that in F_2. (After East.)]

A similar case is shown by a cross between fantail and common pigeons (fig.
86). The latter have twelve feathers in the tail, while the selected race
from which the fantails came had between 28 and 38 feathers in the tail.
The F_1 offspring (forty-one individuals) showed (fig. 87) between 12 and
20 tail feathers, while in F_2 the numbers varied between 12 and 25. Here
one of the grand-parental types reappears in large numbers, while the
extreme of the other grand-parental type did not reappear (in the counts
obtained), although the F_2 number would probably overlap the lower limits
of the race of fantail grandparents had not a selected (surviving) lot been
taken for the figures given in the table.

[Illustration: FIG. 86. Cross of pigeon with normal tail P_1 and fantail
P_1; F_1, bird below.]

[Illustration: FIG. 87. Cross of normal and fantail pigeons. (See Fig. 86.)
The F_2 range is wider than that of F_1. The normal grand-parental type of
12 feathers was recovered in F_2 but the higher numbers characteristic of
fantails were not recovered.]

The preceding account attempts to point out how I should prefer to
interpret the problem of selection in the light of the most recent work on
breeding. But I would give a very incomplete account of the whole situa