Oct. 22 ,1917 Indirect Effects of Selections in Breeding Indian Corn 131 
But on the whole there seems to be some tendency for the upper ears to 
exceed the single ears in length. 
In the circumference of ears the means of the single ears exceed the 
mpatig of the upper ears in 8 out of x 1 cases, but only 4 of these differences 
are significant when judged by probable errors. 
In weight of ears, the means for the single ears exceed those for the 
upper ears in every case but one. Thus, it appears in a general way that 
single ears tend to be relatively heavier and larger in circumference, but 
shorter than upper ears. 
In rows of kernels on ears the mean number for the single ears is larger 
in each case but one than that for the upper and lower ears, but three 
of these differences are so small as hardly to be regarded as significant 
when compared with probable errors. 
differences in variability 
[Table III and fig. 13 to 16] 
With respect to the length of ears nothing very definite is to be in¬ 
ferred. In cir- cumference of ears, with a single exception in 22 cases, 
the standard deviation of the single ears exceeds that of upper and lower 
ears,and the difference is insignificant in this exceptional case. 
In weight of ears the single ears have a greater standard deviation 
than the upper or lower ears in 21 out of 22 cases. In fact, so great is 
this difference that in some cases it looks as if the two frequency distri¬ 
butions of upper and lower ears might be combined to form the frequency 
distribution of single ears. 
In length and weight of ears the coefficients of variability of the 
lower ears exceed those of the corresponding upper ears. In circum¬ 
ference of ears, the coefficient of variability of the lower ears either exceeds 
in each case that of the corresponding upper ears or differs from it by 
an insignificant amount. 
ERECT- AND DECLINING-EAR STRAINS 
Table IV gives the means, standard deviations, coefficients of varia¬ 
bility, and the corresponding probable errors for the characters in ques¬ 
tion of the erect- and declining-ear strains. 
Although we exhibit graphically the results on the erect- and declining- 
ear strains and those on the high- and low-ear strains on the same figures, 
we do so to save space, and not to make comparisons of the erect- and 
declining-ear strains with the high- and low-ear strains. Comparisons are 
made only between the erect- and declining-ear strains and between the 
high- and low-ear strains. 
