FERTILIZATION IN PIMA COTTON. 5 



Table 1.— Hybrids in populations from open-pollinated seed produced by adja- 

 cent rows of Egyptian and upland varieties of cotton at Sacat&n, Ariz., 



in 1920. 



Variety from which seed was obtained. 



Fi hybrids. 



Pima (Egyptian).-. 

 Lone Star (upland) 

 Hold on (upland)... 



That these percentages of hybrids correspond closely to the actual 

 percentages of ovules which were cross-fertilized by pollen of. the 

 other type is indicated by the following facts : The seeds were planted 

 four to the hill and no thinning- was done. Comparison of the per- 

 centages of hybrids in the hills containing one. two. three, and four 

 plants, respectively, showed that while each successive increase in the 

 number of plants was accompanied by a decrease in the percentage of 

 hybrids, the differences were not significant, even as between hills 

 containing one plant and hills containing four plants. Hence, it may 

 be concluded that little, if an}^, natural selection in favor of the 

 hybrid plants had taken place during germination and the seedling- 

 stage of growth. 6 



The difference in the percentages of hybrids between the progenies 

 of the Pima and of the Lone Star plants is apparently not significant, 

 but the percentage of hybrids in the progeny of Holdon is nearly 

 four times as great as in the progeny of Pima, and the difference is 

 7-J times its probable error. So far as this evidence goes, it would 

 seem that when Egyptian and upland cottons are grown in close 

 proximity, the former yields a smaller percentage of vicinists than 

 the latter. It should be noted, however, that during the latter half 

 of the summer the upland plants showed a much greater decline in 

 the rate of flowering than did the Pima plants, and this would favor 

 the production of a higher percentage of upland X Pima than of 

 Pima X upland hybrids. 



The extent of vicinism occurring when upland plants are located 

 in the midst of a field of Egyptian cotton is indicated by the results 

 of an experiment performed at Sacaton in 1920 and 1921. Fifty 

 plants of Acala (upland) cotton were grown in 1920 in the middle sec- 

 tion of the central row of a 7-row plat of Pima cotton. Adjacent to 

 this plat, on both sides, were several other plats which contained only 

 Pima cotton. The rows contained about 400 plants each, so that the 

 upland cotton was completely surrounded by the Egyptian. The 

 arrangement of the planting is shown in Figure 1. 



6 Data given elsewhere in this bulletin indicate that seeds produced by Pima flowers 

 which had been cross-pollinated with upland pollen germinate somewhat better than 

 TimaXPima seeds, although the difference in the germination did not exceed 4.1 ±1 per 

 cent and was therefore too small to affect materially the percentage of hybrids yielded 

 by seed from naturally pollinated flowers on the Pima plants. To illustrate : If the 

 proportion of germination of the seeds resulting from cross-fertilization of Pima ovules 

 with upland pollen was only 4 per cent higher than that of seeds from Pima ovules 

 fertilized with Pima pollen and if the population grown from seed produced by naturally 

 pollinated Tima flowers contained 10 per cent of hvbrids, the actual percentage of 

 hybridized (Pima Xupland) ovules would have been 9.6 per cent. 



