Mar. 20,1916 
Inheritance of Fertility in Swine 
1155 
While it is possible that 90 per cent of the litter sizes in these tables do 
not represent the exact genetic constitution, yet it is probable that in 
general the greater the disparity in litter sizes between the two animals 
in the P generation, the greater will be the expected deviations in the 
F 2 , and the smaller the deviations in the F x generation. The following 
results, Table III, are produced by tabulating the averages of the devia¬ 
tions on this basis. 
Table III .—Average deviations in litter size in the Fj and F 2 generations of swine 
Difference in number of pigs in 
the two P litters. 
0 
1 
2 
3 
4 
5 
6 
1 
F t deviations. 
2.13 
1.89 
i- 93 
2. 04 
2. 19 
I. 82 
I. 32 
I. 12 
F 2 deviations. 
1. 91 
1. 84 
2. 16 
2. 10 
2, l6 
I. 71 
I. 72 
I. 98 
A calculation of the probable errors involved in this table shows that 
only the difference between the F t and F 2 deviations where the disparity 
in litter size is seven pigs is large enough to be mathematically significant. 
The difference when the parents vary from each other by two pigs and by 
six pigs is on the border line between significance and nonsignificance, 
but the five other columns are distinctly unenlightening. Yet, if the 
difference of two pigs is barred, the results are what might be expected. 
One criticism against the preceding method of treatment is thoroughly 
valid. If swine fertility depends on only one or two genetic factors, it 
is obvious that the point at which the difference between the two parents 
occurs is more important than the degree of difference. For example, if 
there is a physiological division point between two hereditary factors at 
six pigs, then a difference of two or even of four below six pigs might not 
be significant, while a difference of one more or one less in a litter of six 
or seven pigs would be thoroughly significant. An examination of the 
data from this point of view is now in progress, but it is probable that the 
key to the situation will only be discovered by breeding experiments. 
CURVES OF LITTER FREQUENCIES 
The distribution of the different sizes of litters in the three generations 
is given in Table IV. 
Table IV .—Litter frequencies in swine 
Generation. 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
11 
12 
i 13 
14 
15 
■p/Expected. 
0.11 
i-5 
9.8 
39 
108 
216 
324 
370 
324 
216 
108 
39 
9.8 
I- 5 
0. it 
Actual. 
3 
9 
30 
80 
124 
198 
300 
362 
3i8 
162 
9i 
59 
26 
6 
3 
p fExperted. 
•05 
•75 
4.9 
19 
54 
108 
162 
185 
162 
108 
54 
19 
4.9 
•75 
• 05 
* M Actual. 
0 
5 
14 
32 
69 
122 
149 
161 
149 
85 
62 
23 
8 
4 
2 
. 
•05 
•75 
4.9 
19 
54 
108 
162 
i8 5 
162 
108 
54 
19 
4.9 
•75 
.05 
r *\Actual. 
0 
4 
17 
32 
63 
107 
154 
172 
135 
95 
59 
30 
11 
3 
3 
