FERTILIZATION IN PIMA COTTON 



23 



occasions at from 3 to 6 p. m., and its reaction in a solution of cane 

 sugar was observed. After immersion during one to five minutes 

 a small percentage of the grains ruptured and their contents oozed 

 out slowly, the phenomenon having been very different from the ex- 

 plosive ejection of a long thread which was observed in pollen 

 grains collected in the morning from open flowers. 



The viability of the pollen of Pima cotton during the day of an- 

 thesis was tested on July 26, 1917, and on July 25 and August 5, 

 1919. As the results show close agreement, only those of August 

 5 will be considered in detail, parallel tests with the Durango va- 

 riety of upland cotton having been made on that date. Pollen 

 of each variety was collected at half-hour intervals from 6 a. m. 

 (hence before the corollas had opened) until 10.30 a. m. and at 

 intervals of one hour thereafter until 3.30 p. m.. with a final col- 

 lection at 5.30 p. m. Shortly before the first collection was made 

 the sky was cloudy, but during the remainder of the day there was 



SO 



&jm7 S 3 /O // /2 /pm. 2 3 

 77Af£: Of £>£~rf/?M/A/s4r/0/V 



Fig. 3. — Indexes of viability of the pollen of Pima (Egyptian) and of Durango (upland) 

 cotton at successive time intervals during the day of anthesis. The curve for Pima 

 is indicated by a solid line and that for Durango by a dotted line. Both curves show 

 a low viability early in the morning, a rapid increase beginning at 8.30 or 9 o'clock, 

 and a gradual decline after midday. 



full sunshine. The tests were made upon one flower of each variety 

 up to 10.30 a. m. and thereafter on two flowers of each variety, 

 the average of the viability indexes of the two flowers having been 

 used in plotting the curves. 



Endeavor was made to select only flowers which were so located 

 on the plant as to have been exposed to full sunlight up to the time 

 of collection. This object was realized in the case of Pima- but not 

 in the case of Durango, owing to the limited number of flowers 

 available. However, no flowers of Durango were taken later in the 

 day which had not been so exposed during several hours, and the 

 earliest flowers in the most exposed situations were selected. The 

 Durango anthers tended to become exhausted of pollen earlier in 

 the day than the Pima anthers, probably because the shallow, flar- 

 ing corolla of upland cotton attracts more of the large pollen-carry- 

 ing insects. 



The indexes of viability of the two varieties at different hours are 

 shown by curves in Figure 3. The percentage of grains ejected, one 



