A COMPARISON OF MAIZE-BREEDING METHODS 3 
Label the rows in the crossed experiment C 1 to C 20 and those of the selfed 
experiment 8 1 to S 20. 
Crossed experiment.—At flowering time, number the plants C 1-1 to C 1-50, 
C 2-1 to C 2-50, ete., select the 10 best-appearing rows, and secure at least two 
ears representing each possible combination of these 10 rows, using best-appear- 
ing plants. Make note on tag of pollen plant used, This will mean at least 90 
SSC te +1)X2. Superior plants to be used as many times as possi- 
‘ble, avoiding reciprocals. 
Determine row yields based on length of ear. Arrange the five highest yield- 
ing rows in order of the yield and select the ears representing crosses 1X2, 1X3, 
1X4, 15, 23,24, 25, 3X4, 3X5, and 4X5 and 10 additional hand-pollinated 
ears representing the highest yielding parent plants without regard to row, thus 
obtaining 20 selected ears for next season’s crossed experiment. 
Selfed experiment.—Obtain at least 5 selfed ears from each of the rows, making 
100 ears. Determine the row yields based on length of ear. From the 100 ears 
‘select a selfed ear from each of the two highest yielding plants of the 5 highest 
yielding or otherwise desirable rows, freedom from diseased or abnormal plants 
being given due consideration, and from the remaining selfed ears select 10 from 
the highest yielding plants, regardless of row, not more than one from each row, 
making 20 selected ears for next season’s selfed experiment. 
PRELIMINARY COMPARISONS 
The first planting was made in 1918, and the procedure described 
above was followed in subsequent years. Beginning in 1920, in ad- 
dition to providing the seed for continuing the experiment, crosses 
were made between selfed lines for preliminary comparisons with 
the crossed experiment. In 1921 F,, or compound F, hybrids, were 
made by crossing the first-generation hybrids made in 1920, and 
_ these compound hybrids were crossed again in 1922, making Fs, or 
double compound hybrids. 
‘The results of these preliminary comparisons are set forth in Table 
1. There is little uniformity in the relative behavior of the various 
seed stocks in the three seasons. The diversity may be due to genetic 
differences, since different selfed lines are represented in the F, hybrids 
of the different seasons. On the other hand, although many of the 
differences appear statistically significant, 1t is not at all impossible 
that the variations are due to errors of sampling. The probable 
errors were calculated from the array of progeny means, taking N = the 
number of progenies. This gives a large probable error, but the 
‘seed stocks compared were not always adjacent in the field, and even 
these large probable errors should not be interpreted too literally. 
‘TaBLE 1.—Comparison of the average ear length of maize in crossbred and selfed 
expervments 
J Number |Number| Mean 
Year and experiment of of length of 
crosses plants ear (cm.) 
— = 
7A Oh 
PLOSSOU LIS TOCKT Peet eee eee Pn ee Wht eles os Ae Lh oF ed 20 933 25. 640. 5 
SEL SOIT VTS 2a TT te | ms eee Oe ee OR Ee Se ee ee, ONE 4 179 34. 1+1.1 
eT UCCTST RSD Sd Ce ee de ee kn 20 496 | 23.34 .8 
“EE DAD™) Ap DSA RO aC A Oa le en 9 205 | 27.1+1.0 
epee Dee eseeres aot cue met) CEP Oe 5 124] 36.2-+ .7 
oes es cvrd, i bel epg ANG IIs lee spd a a. 20 454] 30.44 .6 
Pine) inmates Retiro pai wiih Gren oe oa 16 301 | 33.94 .8 
Jo eT lt AE Si a a Re 12 243.| 26.34 .8 
epee aretha ae 16 298 | 31.04 .7 
