28 
BULLETIN 1468, U. S. DEPARTMENT OF AGRICULTURE 
was present. It was found that the viability of pollen collected at 9 
in the morning was reduced about 50 per cent by being stored for 
7 hours. 
The following season a more extensive program was undertaken 
involving 40 heterozygous plants, but the short-interval pollinations 
were eliminated, the whole effort being placed on a T-hour period of 
storage. 
In practice it was found convenient to collect the pollen, make 
one application of half the total quantity (anther free), and then 
attach the bag containing the remainder to the plant which had 
been selected for pollination 7 hours later. 
Several methods of measuring the quantity of pollen used were 
tried, but as none were satisfactory this refinement was discarded and 
an attempt made to divide the anther-free pollen into two visibly 
equal lots. From the numerous double applications only 13 pairs 
of ears were obtained, many 
<3 
\x> 
Fig. 
/ 
/ 
failing 
on one side or the 
other. The pertinent facts 
with respect to these ears 
are shown in Table 18. 
The immediate interest 
was in the percentage of 
crossing over, and, as the 
table shows, there is little 
difference in the percentage 
of crossing over between 
fresh and old pollen, though 
the average storage period 
is 7 hours and 14 minutes 
and the indicated death rate 
is 35 per cent. 
Although the survival of 
crossover gametes does not 
seem to be affected by the 
factors which brought about 
the loss of viability, there is 
striking evidence that the percentage of pollen grains bearing the. 
waxy character has been greatly increased. 
The data leave no room to doubt that a profound alteration in the 
proportion of waxy to horny pollen grains has taken place. Sim- 
ilarly, though not to the same degree, the proportion of white-carry- 
ing to color-carrying gametes has been altered by the storage of pol- 
len. Analysis of the data, .however, shows that the increased pro- 
portion of white to colored is simply a dependent result of the in- 
crease in the percentage of viable waxy pollen due to the correlation 
between waxy and white. 
Although in pollen stored for seven hours the increase in the per- 
centage of waxy seeds over the expected equality is most striking, 
the fact must not be overlooked that the proportion of waxy seeds 
resulting from the application of fresh pollen is below equality by 
a percentage too large to be attributed to chance. The point is im- 
portant, as it indicates that a simple differential death rate for the 
two sorts of pollen will not account for the observed results. From 
-Inheritance of percentage of crossing 
over, showing progenies of extreme rates 
regressing toward the mean of the parental 
progeny 
