54 
BULLETIN 1468, U. S. DEPARTMENT OF AGRICULTURE 
If changes in temperature or humidity at the time of pollination 
influenced the apparent rate of crossing over there would be de- 
tectable differences in the mean rate of crossing over of ears pol- 
linated at different dates. The mean rate of crossing over for all 
days on which four or more pollinations were made are presented 
in Table 33. Differences in the rate of crossing over in male and 
female gametes and in the presence of heterozygous or homozygous 
R make it necessary to subdivide the ears into four groups and to 
confine comparisons to the ears of the same group. 
Table 33. 
-Crossing over measured on ears of Dh 416LSL1C5L1R23 pollinated 
on different dates 
Female heterozygous 
Male heterozygous 
Date of 
pollination 
RR 
Br 
RR Rr 
Number 
of ears 
■sssr «JS- 
overs | OIears 
Percentage 
of cross- 
overs 
Number 
of ears 
Percentage 
of cross- 
overs 
Number 
of ears 
Percentage 
of cross- 
overs 
Aug. 11 
4 
18. 8il. 51 
5 
5 
10 
19 
8 
17. 4i0. 99 
Aug. 14 
6 
14 
24 
9 
17. 4i6. 74 
20. 3± .62 
18. 3± . 62 
18. fcfc . 54 
18. Si . 77 
Aug. 16 13 
Aug. 20 
19. 9i0. 82 
6 
6 
17. 9± . 49 
21. 3±1. 86 
15. Oil. 30 
17. 6i . 59 
Aug. 23 
22. 2± .94 
18. Oil. 14 
The most significant difference is that between August 14 and 16 
in the R R group of male heterozygous. The difference is 2.9 ±0.97 
per cent, which is three times the probable error. This departure 
is not too large to be ascribed to chance, for the difference is the 
largest of 20 possible comparisons. The magnitude, however, is of 
the same order as the variations significantly associated with other 
factors. The absence of any significant correlation between cross- 
ing over and the date of pollination is evidence that there is no pro- 
gressive change in the rate of crossing over as the season advances, 
but daily fluctuations would not be reflected in this correlation. 
In seeking an explanation for differences between families in the 
rate of crossing over of G and Wx, Stadler (16) considered differ- 
ences in flowering dates and concluded that there was no apparent 
relation of flowering dates to crossover percentages. 
Another method of detecting daily fluctuations is to compare the 
mean standard deviation of crossing over on individual days with 
the standard deviation of the entire population. If the mean rates 
of crossing over on individual days differ among themselves more 
widely than should follow as the result of random sampling, the 
mean standard deviation on individual days should be less than that 
for the entire population. The mean standard deviation of crossing 
over of ears pollinated on the same day is 4.81 ±0.19, and that for the 
entire population is 4.57 ±0.16. The results indicate that the crossing 
over of ears pollinated on individual days were random samples 
from the general population and that changes in the meteorological 
conditions at the time of pollination did not influence the apparent 
rate of crossing over. 
