As part of a progran of hybrid tomato seed production at the University of California College of Agriculture at Davis, the longevity of tomato pollen was tested under various storage conditions. Pollen was collected by means of a buzzer device similar to that described by Cottrell-Dormer (1945). In a few minutes this collector can collect enough pollen to supply a worker all day. The question naturally arose whether pollen could be safely used all day, or perhaps longer.
A preliminary test using the parents involved in the hybrid cross Pearson (ms2) x Pennheart was made.
It was found that pollen stored in a open vial in the full midsummer sun can still set a fruits on the third afternoon after the morning of collection. No fruit was set later by this pollen.
Other lots of pollen, stored under various conditions, produced fruit a much longer period. That stored over CaCl2 in a refrigerator produced fruit on 50% of flowers pollinated(five out of ten flowers) - )35 days after collecticn.
A more careful test was made the following year, using as female plants San Marzano (ms9) and as pollen parents a double-haploid line of San Marzano (2.72).
A quantity of pollen was collected three months before the start of the test, in order to lengthen the effective time span of the experiment. Each of the subsequent batches of pollen was collected on a single mourning andthoroughly mixed before dividing into storage lots. Three such collections were made, six days apart, the collections being kept separate in storage. This was done to reduce the effect of random fluctuations of weather by summing results of each storage condition at given pollen ages.
Storage temperatures were 0 deg., 10 deg. and 20 deg. C. At each temperature one lot of pollen was stored in loosely-capped vial, another in a sealed vial containing CaCl2. Control pollinations with fresh pollen were made each time any experimental pollinations were made. Precautions were taken te reduce the effect of variation between female plants.
Pollen viability was measured as percent of flowers that set fruit, and as number of seed per fruit. It may here be noted that this is an absolute measure of the effectiveness of pollen, not achievable in germination tests in vitro. Ten flowers were pollinated at each treatment with each lot of pollen. Pollinations were at three-day, intervals for a month, then weekly for two months, then monthly until the oldest pollen was a year old.
At each temperature pollen stored in low humidity (over CaCl2) produced fruit and seed long after that in high humidity (loosely capped vial). The lower the temerature of storage, the higher the life of the pollen. All samples retained their ability to stain in acetocarmine reguardless of their capacity to produce fruit.
Decline of ability of a given sample to set fruit and seed seemed parallel, though many samples set a few parthenocarpic fruits before complete failure.
Pollen collected during a period of unfavorable weather (hot and dry wind) had a much reduced viability both initially and in storage life, which was comparatively short. Ovules (or ovaries) also showed reduced fertility during unfavorable weather.
Pollen stored under the best conditions of the experiment (O deg. C. and low humidity) produced fruit after a year in storage but the practical limit for seed production appears to be six months.
Using these conditions onion pollen has been viable after two months, and pollen of Lycopersicon peruvianum over three months.