14 



BULLETIN 1134, U. S. DEPARTMENT OF AGRICULTURE. 



Table 7. — Measurements of the reproductive organs in Pima and Holdon 



cottons. 



Variety. 



Pima (Egyptian) . 

 Holdon (upland). 



Differences. 



Number 



of plants 



on which 



flowers 



were 



ured. 



Number 

 of flowers 



ured. 



Mean length of parts (millimeters). 



Stamens. 1 



5.0±0.01 

 8.0± .03 



3.0± .03 



Pistil.* 



36. 3± 0.29 

 26. 5± .14 



9.8± .32 



Stigmas. 



9.8±0.21 

 3.1± .08 



6.7± .22 



Relative 

 length of 

 stigmas 

 (percentage 

 of pistil 

 length). 



27.0 ±0.45 

 11. 7± .48 



15. 3± .65 



1 Measured from the base of the filament to the apex of the anther. The average length of 10 stamens 

 per flower was taken as the unit in computing these means. Stamens from near the middle of the stamina 1 

 column were measured. The uppermost stamens are much shorter. Measurement in 1921 of 5 upper 

 and 5 middle stamens per flower in 5 flowers, each from a different plant of Pima cotton, gave the following 

 means. Uppermost stamens, 2.4±0.04 millimeters; middle stamens, 4.3±0.05 millimeters. 



1 Measured from the bottom of the corolla (hence somewhat below the summit of the ovary) to the apex 

 of the stigmas. 



Examination in 1920 of the flowers of 20 varieties of upland cotton 

 growing at Sacaton, Ariz., showed that, shortly after the corolla 

 begins to open, the upper stamens are erect. In 16 varieties they 

 were from slightly shorter than to slightly longer than the stigmas, 

 while in 4 varieties the stigmas exceeded the stamens by lengths not 

 greater than 5 millimeters. 11 In most of the upland varieties, there- 

 fore, the whole or the greater part of the length of the stigmas is 

 surrounded by the stamens, and the erect position of the latter during 

 the first hour or so after the opening of the flower brings the anthers 

 into contact with the stigmatic surface. Unlike the condition in 

 Egyptian cotton, there is a limited power of movement, for later in 

 the day the filaments become more nearly horizontal. It was ob- 

 served in 1921 that in the Acala variety at 3 p. m. most of the upper- 

 most filaments diverged at angles of 20° to 45°, although even at this 

 hour occasional anthers remained in contact with the stigmas. The 

 entire length of the stigmas in upland cottons is at all times, how- 

 ever, much more accessible to foreign pollen than is the interstamen 

 section of the Egyptian stigmas. 



If one overlooks the receptive character of the entire surface of 

 the pistil outside the stamina! sheath and the possibility of a high 

 degree of fertilization by self-pollen automatically discharged upon 

 the basal portion of the stigmas, the assumption is likely to be made 

 that varieties of cotton having long stigmas are not well adapted to 

 self-fertilization. Meade (40) drew this conclusion from the results 

 of an experiment with upland cottons performed by him at San 

 Antonio, Tex., in which flowers of a short-style variety (Acala) and 

 of a longer styled variety (Durango) were artificially pollinated, the 

 stigmas having been thoroughly smeared with their own pollen. 

 Comparing these artificially pollinated flowers with naturally pol- 

 linated flowers of the same varieties in respect to the percentages of 

 bolls set, the variety having long stigmas showed a mean increase 

 of 11.0 ±2.2 per cent from artificial pollination, as compared with an 

 increase of only 5.3 ±2.4 per cent in the variety having short stigmas, 



11 It is possible that the stigmas were abnormally short in some of the upland cottons 

 grown at Sacaton in 1920, among which were several of the long-staple varieties. Accord- 

 ing to Meade (1/0) in mauy of the long-staple upland varieties the stigmas often exceed 

 the anthers by 15 millimeters. 



