24 BULLETIN 1134. U. 5. DEPAETMEXT OE AGRICULTURE. 



of the factors in computing the index, was merely estimated, except 

 in a few cases when the total number of pollen grains in the field of 

 the microscope was small. The more rapid decline in viability during 

 the afternoon shown by the Pima pollen was probably caused by the 

 fact, already noted, that the flowers of this variety were more ex- 

 posed to the direct rays of the sun than the Durango flowers. 



Pollen has been found to retain its viability much longer in flow- 

 ers which have been inclosed in paper bags to prevent cross-pollina- 

 tion than in open flowers. On July 26. 1917. buds due to open that 

 morning were bagged at 6 a. m., and the viability of the pollen was 

 tested in sugar solution at 6 p. m. of the same day and at 6 a. m. 

 of the day following. The results are given in Table 11. 



Table 11. — Prolongation of the viability of the pollen in cotton flowers bagged 



at 6 a. m. July 26. 



After immersion 

 until ejection— ___. 



Number Esti- viability 



Hour of testing. of ; mated Yifi^r 



tfegan. actively> 



index. 



(5 p.m., Julv 26 



Seconds. 

 2 37 



Seconds. 

 105 

 270 



Per cent. 1 

 75 | 

 40 ! 



71 



6a.m..' Julv 27 



1 35 



15 









Whereas in the case of unbagged flowers the proportional ejec- 

 tion of pollen at 6 p. m. of the day of anthesis was estimated at 

 only 3 or 4 per cent, pollen from bagged flowers at the same hour 

 eiected with great vigor, and the percentage of grains ejected was 

 almost as high as in the case of pollen from uninclosed flowers shortly 

 before noon of the day of anthesis. At 6 a. m. of the day following 

 anthesis the pollen from bagged flowers ejected more slowly and 

 less completely. It seems probable, therefore, that even in bagged 

 flowers the pollen loses its viability during the day following 

 anthesis. 15 



It may be deduced from the curves shown in Figure 3. which are 

 based upon an index integrating the percentage of pollen grains 

 which eject their contents and the rapidity with which ejection is 

 completed, that under conditions at Sacaton. Ariz., the viability of 

 the pollen of Pima (Egyptian) and of Durango (upland) cotton is 

 low during the early morning hours, begins to increase rapidly at 

 about 9 o'clock, and begins to decline at or shortly after midday. If 

 the index of viability based upon the rapidity and completeness of 

 ejection. in a sugar solution indicates the capacity for normal germi- 

 nation, it would be concluded that pollen which reaches the stigmas 

 before 8 or 9 a. m. will germinate more slowly and less completely 

 than pollen which arrives later in the morning. It should be noted, 

 however, that while at earlier hours a much longer time was re- 

 quired for the ejection to take place, the percentages of the total 

 number of grains which finally ejected their contents were in some 

 cases relatively high. Thus, in the case of Pima cotton, ejection 



15 Pollen, longevity in the snapdragon and in maize is the subject of a recent publication 

 bv H. E. Knowlton* [SB), who summarizes (p. 755-759) the results of other investigators 

 with various plants and points out (p. 7S6) that pollen may retain its capacity to germi- 

 nate when no longer able to effect fertilization. 



