FERTILIZATION IN PIMA COTTON. 25 



finally took place in about TO per cent of the grains collected at 

 6.30 a. m. on July 26, 1917, and in about 65 per cent of those col- 

 lected at 6 a. m. on July 25, 1919. A test of Durango pollen on 

 August 5, 1919, showed ejection at 6.30 a. m. in about 75 per cent 

 of the total number of grains. It also seems probable that pollen 

 discharged at an early hour may continue to mature after it has 

 reached the stigmas. 



DEGREE OF MATURITY OF POLLEN AS AFFECTING FERTILIZATION. 



An experiment was performed in 1921 to ascertain whether fer- 

 tilization can be effected by immature pollen placed upon the stig- 

 mas many hours in advance of the time of anthesis. Pima flower 

 buds were emasculated in the evening, pollen squeezed from the an- 

 thers of the same flower was placed upon the stigmas, and the flowers 

 were kept inclosed in bags until the stigmas had withered and there 

 was no longer danger of accidental cross-fertilization. Only 1 of 25 

 flowers thus treated produced a boll which reached maturity. This 

 boll contained five ripe seeds. A second boll was retained longer 

 than 10 days but finally dropped, the exact date not having been 

 ascertained. The remaining 23 flowers shed their undeveloped bolls 

 within 10 days of the date of pollination, this being apparently the 

 average number of days from anthesis to shedding for Pima cotton 

 in Arizona (SI), p. 15). The results of this experiment are of prac- 

 tical interest as showing that when flowers are emasculated the even- 

 ing before anthesis for the purpose of making hybrids, every precau- 

 tion should be used to prevent self pollen from reaching the stigmas, 

 fertilization with such pollen being possible, although evidently not 

 frequent. 



The methods used in emasculating and bagging flowers in this 

 and in experiments subsequently described are illustrated in Plates 

 IV and V. 



In order to ascertain whether fertilization is affected by deferring 

 pollination several hours after it would take place normally, 240 

 Pima flower buds were emasculated late in the afternoon preceding 

 anthesis during the period from July 22 to August 2, 1921, 20 buds 

 having been treated on each day of the experiment. Other flower 

 buds were bagged at the same time to supply the pollen required. 

 Half of the flowers were pollinated at 8 o'clock the following morn- 

 ing and the others at about 5 p. m. In open flowers the anthers 

 would have been practically empty of pollen, and the stigmas would 

 have been losing their turgidity at the latter hour, but it has been 

 shown that bagging tends to prolong the freshness of the flower. 



The number of bolls which matured and the number of seeds in 

 each boll were determined, and from these data were computed the 

 percentage of bolls matured and the mean number of seeds per boll, 

 as stated in Table 12. The difference in the percentage of bolls 

 matured was in favor of the deferred pollination, but was less than 

 three times the probable error. On the other hand, early pollination 

 yielded a somewhat higher mean number of seeds per boll, and this 

 difference was approximately three times its probable error. It may 

 be concluded that in bagged flowers the pollen retains its ability to 

 effect fertilization practically unimpaired up to 5 p. m. of the day 

 of anthesis. 



22421—23 4 



