FERTILIZATION IN PIMA COTTON. 11 



blossom, yielded approximately equal percentages of hybrids points 

 to the conclusion that the higher percentages of vicinists usually 

 obtained from seeds produced by upland plants than from seeds 

 produced by Egyptian plants when the two types are grown side by 

 side is due partly to the earlier slowing down of the rate of flowering 

 in the case of upland cotton. Evidence presented in another part of 

 this bulletin indicates, however, that there may be an intrinsic differ- 

 ence in the liability to cross-fertilization of the two types. 



CONCLUSIONS REGARDING THE PREPONDERANCE OF SELF-FERTILIZATION. 



There is much variation in the percentages of vicinists, or natural 

 hybrids, formed when two distinct types of cotton are grown in prox- 

 imity, as is shown by the results obtained by other investigators and 

 by the writer. This is to be expected in view of the many variable 

 factors involved, such as local differences in the number and kind of 

 pollinating insects and differences in the habit of growth and period 

 of flowering of the varieties. The proportion of vicinists rarely ex- 

 ceeds 20 per cent, however, and is usually much smaller. The avail- 

 able information in regard to vicinism therefore points strongly to 

 the conclusion that in cotton self-fertilization greatly predominates 

 over cross-fertilization. It should not be inferred, however, that 

 because most of the ovules normally are self-fertilized, such cross- 

 fertilization as occurs is negligible in its effect upon the uniformity 

 of a variety. 



4 As a rule, the percentage of vicinists decreases rapidly as the 

 distance between the seed-bearing and the pollen-bearing parents 

 increases, but the data at hand do not permit a conclusion to be drawn 

 as to the degree of isolation necessary to eliminate the danger of 

 cross-pollination. This is doubtless affected by the nature of the 

 varieties grown, by local and seasonal variations in the insect popu- 

 lation and in the flowering of other plants, and by topography, 

 weather, and other factors. 



The percentage of recognizable vicinists produced under natural 

 conditions does not measure the proportion of cross-fertilization 

 occurring, for the reason that many of the ovules are cross-fertilized 

 by pollen from other plants of the same variety. An experiment was 

 performed at Sacaton, Ariz., in which this source of error was 

 eliminated by growing scattered plants of one type in a field of an- 

 other type and allowing only one flower to open daily on each of the 

 isolated plants, seed from which was planted the following season. 

 Plants thus treated yielded 12 per cent of hybrids in the case of 

 Pima (Egyptian) and 28 per cent in the case of Acala (upland). 

 It is believed that these percentages correspond very closely to the 

 percentages of cross-fertilized ovules. 



The results of this experiment indicated that in Pima 88 per cent 

 and in Acala 72 per cent of the ovules were self- fertilized. Other 

 evidence has been obtained at Sacaton that upland X Egyptian are 

 more numerous than Egyptian X upland vicinists. That this is 

 due partly to an earlier decline in the flowering rate of upland as 

 compared with Egyptian cotton is suggested by the fact that seeds 

 produced by flowers of Pima and of Acala cotton which opened 

 during a period when both types were blossoming freely yielded 

 approximately the same percentage of vicinists. 



