638 MINNESOTA BOTANICAL STUDIES. 



The other part of this experiment is to find about the same 

 facts for the pollen. I found beyond doubt that it is not effec- 

 tive so soon as the stigma of the flower will hold the pollen- 

 grains. To ascertain this, I castrated several flowers just 

 opening and used their pollen to pollinate other castrated 

 flowers. I also pollinated the former castrated flowers with 

 mature pollen. At this stage the pollen-grains were not effec- 

 tive in a single instance. But the stigmas were acted on by 

 the mature pollen, producing fruit. To ascertain how long 

 the pollen retains its vitality, I placed some in a box and 

 pollinated castrated flowers with it for seven days in succes- 

 sion. It had lost none of its potency on the seventh day. 

 Seven days is surely long enough to give every chance for 

 cross-pollination by the pollen carried by the bees, and also 

 the pollen may remain on the stigma seven days while the 

 latter is developing if necessary. After the seventh day the 

 pollen begins to lose its vitality and after the fourteenth will 

 not germinate. 



Another experiment was the study of the anthers to ascer- 

 tain when and how the pollen sacs shed the pollen. When the 

 corolla begins to open the stamens are of a greenish-yellow 

 color, and the sacs are completely closed. In about two days 

 the stamens have assumed a bright yellow color, and the sacs 

 are beginning to open. The best time to get pollen for. arti- 

 fical pollination is soon after or just before the sacs begin to 

 open, when it is easily scraped out on a knife-blade and trans- 

 ferred to the stigma of another flower. The average time re- 

 quired for sacs to begin to open after the corolla has started to 

 expand is two days, but there is a considerable amount of varia- 

 tion in the time due to the size of the flower and to the hygro- 

 metric state of the atmosphere. The pollen-sacs open sooner 

 in small than in large flowers, and sooner in dry than in wet 

 weather. In very dry weather while castrating small flowers 

 as those of the Red Cherry tomato, or the Yellow Plum tomato, 

 as soon as the corolla has begun to open, the anthers occasion- 

 ally snapped open, and the pollen could be seen flying about 

 like dust. It will be seen that the safest time to castrate is on 

 a rainy day or while the dew is on, and that the best time to 

 pollinate artificially is in the dryest part of the day. 



Seventy-five flowers were castrated and left to the influence 

 of wind and insects. Some were castrated just as the calyx 

 was beginning to open, others as soon as the corolla had begun 

 to open, and the rest after the latter had fully expanded. All 



