PRODUCTIVENESS OF SELF-FERTILIZED CORN 15 
It should be possible to eliminate from further experiments many 
of the lines used, thus reducing the number of crosses to be compared 
in determining the most productive single and double crosses. It 
should also be possible to select certain lines for crossing with a reason- 
able assurance that the combinations will be materially more pro- 
ductive than the open-fertilized variety. Thus, 2-2-1-3-2-1, 
2-2-2-4-6-2, both of the 5-1-3- lines, 7-4-1-7-4-1, 7-4-2-6-2-2, 
and both of the 10-3-2- lines may be eliminated without any apparent 
danger of serious loss. The 36 crosses necessary to obtain all desired 
combinations between the remaining 10 lines then can be compared 
with greater accuracy. At the same time combinations between 
crosses may be obtained to determine the better double crosses. 
DISCUSSION 
Omitting from consideration the cross grown in field row 64d 
because of the few plants, the average yield of the first six crosses 
listed in Table 6 exceeded that of their checks by 1.57 pounds, or 30 
per cent. That all of these crosses involved a line of the 10-3-1- 
strain and that three of them were between 7-4-1-6-6-2 and a 
10-3-1- line is excellent evidence that the superior productiveness 
of these particular crosses was not due to chance. There can be 
no question, then, that significantly larger yields have been obtained 
by the methods followed. Whether larger yields can be obtained 
practically remains to be proved by further experiments, although 
indirect evidence indicates that they can. 
The strain cross 1 0-3- X 7-4- yielded about 6 per cent more 
than No. 201, F 6 , in the 1921 experiments (6). The F t of the varietal 
cross, WhatleyxSt. Charles White, yielded about 9 per cent more 
than No. 201, F 6 , or about 3 per cent more than 10-3- X 7-4- in the 
same experiments. Other strain crosses yielded slightly more than 
1 0-3- X 7-4-, but none were significantly more productive than 
WhatleyxSt. Charles White, from which No. 201 originated. On 
the basis of these results it was concluded that "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 ¥ t 
varietal cross each year" (6, p. 19). 
The average yield of the 11 crosses between lines of the 7-4- and 
10-3- strains again was about 6 per cent more than No. 201 in the 
1923 experiments. This average superiority, however, was com- 
posed of individual yields, some superior by much more than 6 per 
cent, together with others that were inferior to the variety. The 
average yield of the three better crosses between 7-4-1-6-6-2 and 
lines of 10-3-1- was 1.40 pounds, or 30 per cent, more than the 
yield of No. 201 in the corresponding checks. On the basis of the 
9 per cent superiority of the WhatleyxSt. Charles, F 1; over the 
Delta Prolific, as previously reported (6) , this is an indicated superi- 
ority of 21 per cent for these three crosses over the F ± varietal cross. 
It is recognized clearly that such a detailed conclusion is unwarranted 
on the basis of data from two experiments differing in time, space, 
and method. The results of the experiments in 1921 and 1923 are 
consistent, however, and the indicated difference is large. The more 
general conclusion that some of the crosses produced significantly 
larger yields in 1923 than could have been obtained from the F t 
