SEED-EAR CHARACTERS AND PRODUCTIVENESS IN CORN Vé 
data (4) on the greater smut susceptibility, barrenness, chlorophyl 
abnormalities, and general undesirability in selfed strains with the 
relatively larger numbers of rows are highly suggestive in this con- 
nection. If few-rowedness be looked upon as a more primitive, less 
specialized condition, it is easy to understand why the more com- 
plex mechanism should break down more frequently. Unfortunately, 
the basic data in the present case were not suitable for differentiat- 
ing between the effects of kernel size and number of rows. ‘This 
possible negative relation between yield and number of rows per se 
is one that should be kept in mind. | 
There is nothing in the data to indicate that selection on the basis 
of seed-ear characters could be used_as a method of breeding. At 
the same time, the tendencies shown are of interest in connection 
with the mass selection of seed corn. The ears studied represented 
a carefully selected population, already above the average of the 
varieties in length, circumference, and weight. The positive corre- 
lations between yield and weight of ear in this selected population 
indicate that the selection of larger ears for seed is worth while. 
They also seem to warrant the recommendation to select the longer, 
heavier ears with proportionately heavy cobs and with relatively 
few rows of wide, thick kernels. This is in substantial agreement 
with most of the results of comparisons made by other investiga- 
tors (7). Larger butt circumference seems more desirable if unac- 
companied by an increase in the number of rows, but whether it is 
desirable in any event appears to depend upon the variety.- It 
should be noted that, with the possible exception of number of 
kernel rows, the relations indicated are of size rather than form and 
that the kind of ear indicated as better may be described as a large 
ear with large kernels and a proportionately heavy cob. 
SUMMARY 
Data from ear-to-row plats in four varieties of corn, comprising 
3,265 ears in all and extending over 47 crop years, were studied as to 
possible relations between productiveness and some physical charac- 
ters of the seed ears. 
(1) Accidental variation in soil and experimental conditions was 
eee for a large part, possibly 90 per cent, of the total variation 
in yield. : 
(5) From 2.5 to 6.7 per cent of the total variation in yield in the 
different varieties was a function of variation in the ear characters 
studied. 
(3) On the basis of the multiple regression equations involving 
seven ear characters, yield was related positively to weight of ear and 
length of ear and negatively to number of rows and number of kernels 
per row in each of the four varieties. Similarly, yield was related 
positively to butt circumference and weight of cob and negatively to 
tip circumference in each of three varieties, the relations being reversed 
in the fourth variety. 
(4) The use of weight of ear, number of rows, and number of kernels 
per row in a multiple regression equation gave almost as good a pre 
dicting equation in each of the varieties as the inclusion of the four 
other characters and was the best general predicting equation inyolvy- 
ing only three characters. 
