EFFECTS OF SELECTION ON THE YIELD OF CORN. 19 
CONCLUSIONS. 
The Fj generation of the cross Whatley X St. Charles White was 
markedly more productive than the average of its parents, yielding 
slightly more than its better parent. Whatley. Open-fertilized mass 
selection was ineffective in maintaining this high productiveness, the 
yields of the later generations approximating the average of the 
parents. This is what would be expected following random mating 
if heterosis is due to the complementary action of dominant growth 
factors. It was necessary to select the earlier segregates, as well as 
those having larger ears, because of practical considerations. Selec- 
tion, therefore, was away from the type of Whatley. the parent which 
was higher in yield, but which was undesirable because of its lateness 
and small ears. On the other hand, the F t generation produced as 
much as Whatley and matured slightly, if at all, later than St. Charles 
White. The value of the F 1 generation, therefore, was in its increased 
efficiency in production rather than in an actual increased yield, 
Whatley being taken as the criterion. 
Although it was impossible to retain the efficiency of the F x gener- 
ation of the cross, the method of crossing followed by mass selection 
has established a variety of economic value for the locality. Xo. 201 
has a good length of season, well-proportioned plants, and ears of 
good quality. It has been grown on a commercial scale for several 
years with good results, several hundred acres having been grown 
in 1921. In the present experiments as well as in others it has yielded 
somewhat more than St. Charles White and shown itself to be among 
the higher yielding varieties for this section. It is not, however, the 
outstanding corn that it was in the F x generation. 
Strains have been isolated by selection within self-fertilized lines 
which when intercrossed have produced as much as the original F l 
varietal cross in a single season's comparison. Whether these crosses 
can repeat or not and whether better ones can be found remains to 
be determined. So far there has been no advantage in yield from the 
laborious methods of hand pollinating over what could have been 
obtained by growing the F t varietal cross each year. Selection has 
been carried on for only four years and on a somewhat limited scale. 
however, and it seems reasonable to hope for higher yielding crosses 
when further combinations have been tested. 
Of necessity, selection so far has been rather a haphazard affair, 
based on assumptions that may or may not be founded on fact. 
The results are promising, therefore, in the tendency shown by certain 
strains to give high-yielding crosses. The behavior of strain 10-3- 
when selfed and when crossed is of particular interest in this connec- 
tion. Two possibilities already have suggested themselves in explana- 
tion of the value of this strain in combination, and. it is hoped that 
investigation will determine what the controlling conditions are. 
With the solution of this and similar problems as they present 
themselves from time to time, it should be possible to formulate 
principles of selection based upon fact. Until that time, however, 
reliance must be placed upon empirical methods, and the best chances 
for success seem to lie in the use of large numbers and in avoiding 
preconceived notions as to what are the best pure lines. 
