VARIABILITY IN LINKAGE OF CHARACTERS OF MAIZE 
45 
from a low percentage of the crossover class on a single ear having 
but 61 seeds. Although the probable error is large, the percentage 
departs from the mean of the ears of this group by 4.3 times the 
error, which is rather large to be attributed to errors of random 
sampling, and there would seem to be some justification for discard- 
ing this pair of ears from the data under consideration. 
The criterion for the rejection of outlying observations proposed 
by Irwin (10) gives a value for P of less than 0.001. This means 
that the odds against these observations being a chance departure 
are more than 1,000 to 1. 
If this pair of ears is rejected the mean difference becomes 
— 1.696 ±0.38, a difference almost 4.5 times the probable error. The 
magnitude of the difference, as well as its significance, approaches 
and is comparable with those observed where linkages are present. 
Such results are disconcerting and tend to discredit the significance 
of differences observed in the actual rate of crossing over in the two 
sexes, at least in so far as these differences are attributed to linkage 
relations. 
RATE OF CROSSING OVER NOT ALIKE IN THE UPPER AND LOWER EARS OF THE 
SAME PLANT 
Having found differences in the apparent rate of crossing over 
between the male and the female, it seemed desirable to carry the 
analysis a stage further and examine the data for possible differences 
between upper and lower ears. If the observed differences between 
male and female were due to differences in environment at the time 
of the maturation divisions, it seemed not unreasonable that the time 
elapsing between the divisions in the first and second ears would pro- 
vide the possibility of a changed environment. To a certain extent 
upper and lower ears correspond to the broods in Drosophila, where 
Bridges (3) has shown that the rate of crossing over in the second 
chromosome is lower in the second broods than in the first. 
Table 27. — Crossing over in upper and lower ears of progeny Dh 416L3L1C5L3- 
L1R23 where mole gametes are measured and both ears pollinated at the 
same time 
Upper ear 
Lower ear 
Difference 
(lower— 
upper) 
Parent plant designation 
Number 
of 
seeds 
Percent- 
age of 
crossover 
Number 
of 
seeds 
Percent- 
age of 
crossover 
9 <? 
5702X5905 
379 
170 
660 
598 
516 
564 
525 
266 
26. 1±1. 5 
12. 9±1. 7 
22. 3±1. 1 
17. 9±1. 1 
16. 3±1. 1 
15. 3±1. 
18. 5±1. 1 
12. Oil. 3 
82 
340 
593 
583 
566 
703 
568 
28. 0±3. 3 
17. 6±1. 4 
23. 4±1. 2 
21. 4±1. 1 
16. Oil. 
22. Oil. 
21. 4il. 2 
1. 9 i3. 7 
5710X5909 
4. 7 i2. 2 
5725X5931 
1. 1 il. 6 
5759X5912 . .... 
3. 5 il. 6 
5718X5903 -. 
-.3 ±1.5 
5743X5950a 
6. 7 ±1. 5 
5781X5972 
2. 9 ±1. 6 
5795X5991 
489 16. 8±1. 1 
4. 8 ±1. 8 
3,678 
3,924 
3. 23± . 55 
If the male gametes are subject to a selective force, it is within the 
realm of possibilities that the environment afforded by second ears 
would differ sufficiently from that of the upper ears to bring about 
differences in the survival of the gametic classes. 
