FERTILIZATION IN PIMA COTTON. 



39 



family which has been strictly inbred (selfed) during seven genera- 

 tions, were used in this experiment. The treatments were as follows : 



(A) Flower emasculated and bagged the evening before anthesis and 

 pollinated the following morning with pollen from another flower on the same 

 plant. 



(B) Treatment similar to the above except that pollen from other plants of 

 the same variety was used, these having been of the open-pollinated stock, not 

 of the inbred family. 



The use in treatment A of pollen from another flower on the same 

 plant insured self-fertilization unless somatic variation had occurred, 

 and of this there was no indication. 25 The possibility that the pollina- 

 tion might have been less thorough in the self-pollinated than in the 

 cross-pollinated flowers also was eliminated by this method. In addi- 

 tion to the percentages of bolls matured and the mean numbers of 

 seeds per boll, determinations were made of the mean weights and 

 the germination percentages of the seeds resulting from the two 

 treatments. The data of this experiment are given in Table 20. 



Table 20. — Comparison of the results of self-pollination and of cross-pollina- 

 tion within the variety in an open-pollinated stock of Pima cotton and in a 

 family which had been closely inbred during seven generations. 



[All flowers emasculated.] 



Population. 



Pollina- ! Flowers ! Pe n r f c ^ g < nuSber 

 tion. treated. | J^^l I of seeds 

 per boll. 



1 



O pen-pollinated i Self : 



Do Cross 



Difference. 



Inbred. 

 Do. 



Self. 



Difference. 



weight 

 of 100 



165 

 162 



Percentage 

 of germi- 

 nation. 



91. Oil. 5 ; 17.0±0.17 

 85.8±1.8 ! 16. 7± .18 



5.2±2.3 , .3± .25 



12.8±0.04 

 12. 6± .08 



.2± .09 



155 i 92.9±1.3 

 151 i 86.3±1.7 



17. 2± .16 

 17. 5± .18 



6.6±2.1 .3± .24 



,9± .09 



90.8±0.9 

 88.2±1.0 



2.6±1.3 



13. 5± .05 i 86.8±1.0 

 12.6± .08 ! 89.7± .9 



2.9±1.3 



The data given in Table 20 show little difference in the results of 

 the two treatments in either population. The only differences that 

 appear to be significant occurred in the inbred population in re- 

 spect to the percentage of bolls matured and the mean weight of 

 seeds, self-pollination having given the higher value in both cases. 

 In the mean number of seeds, the real criterion of the relative com- 

 pleteness of fertilization, neither population showed a significant dif- 

 ference. The outcome of this experiment warrants the conclusion 

 that within the Pima variety there is practically no difference in 

 compatibility between self pollen and pollen from other plants. 



2B Emoto (16) tested species of Primula. Brassica, Hyacinthus, Preesia, etc., as to the 

 comparative effects of autogamy (fertilization by pollen of the same flower), geitonogamy 

 fertilization by pollen from another flower of the same plant), and xenogamy (fertili- 

 zation by pollen from another plant). The criteria used were fruitfulness, length, and 

 width of the capsules, number of seeds per capsule, weight of seeds, and germination of 

 seeds. This author concluded that geitonogamy was superior to autogamy in very few 

 cases. Darwin (IS, p. 329) concluded from the results of his experiments with plants 

 belonging to numerous families that " in very few cases did crossing different flowers of 

 tne same plant as compared with selfing a flower with its own pollen appreciably in- 

 crease the number of seeds produced." 



