journal of fflram research 
Vol. XXVII Washington, D. C., February 9, 1924 No. 6 
SELECTIVE FERTILIZATION IN COTTON 1 
By Thomas H. Kearney, Physiologist in Charge, and George J. Harrison, Field 
Assistant , Alkali and Drought . Resistant Plant Investigations , Bureau of Plant 
Industry , United States Department of Agriculture 
INTRODUCTION 
It has been shown that in cotton, under natural conditions of polli¬ 
nation, a large majority of the ovules are self-fertilized, although the flower 
is well adapted to cross-pollination and foreign pollen usually reaches 
the stigmas (7, 2 p. 9, 10, 34, 55). There is also evidence of complete 
compatibility, even between different species such as Upland and Egyp¬ 
tian cottons, for pollen of both species proved to be equally effective in 
accomplishing fertilization when applied separately to the stigmas of 
Egyptian cotton flowers (7, p. 40-42). The question therefore suggests 
itself whether selective fertilization, in favor of the like pollen when 
unlike pollen also is present, is a factor in the observed predominance 
of self-fertilization. 
The results of experiments by Balls, by McLachlan and by Kearney, 
summarized in another paper (7, p. 42-49) showed that selective fertili¬ 
zation does occur in cotton. When the stigmas of Egyptian cotton were 
pollinated simultaneously with approximately equal quantities of Egyp¬ 
tian and of Upland pollen, a large majority of the ovules were found to 
have been fertilized by the like pollen. The percentage of Egyptian X 
Upland hybrids resulting from such double pollinations was 10.8 ± 0.6 
in an experiment performed by Argyle McLachlan and 18.4 ± 1.5 in an 
experiment conducted by one of the writers, whereas, if both pollens 
had been equally effective in accomplishing fertilization, 50 per cent of 
the resulting plants should have been hybrids. 
Yet no general conclusion could be drawn as to the importance of 
selective fertilization in cotton, because experiments made hitherto in 
Arizona have not afforded decisive evidence that it occurs in the Upland 
type also. Reciprocal double-pollinations on Upland cotton by 
McLachlan indicated little or no selective fertilization, the percentage 
of Upland X Egyptian hybrids having been 42.3 ± 1.1. Earlier experi¬ 
ments at Sacaton, Arizona, to determine whether selective fertilization 
takes place in Upland cotton failed because nearly all of the artificially 
pollinated flowers on the Upland plants were lost by abscission. 
The writers, assisted by Max Willett and Dow D. Porter, have now 
succeeded in obtaining conclusive evidence that selective fertilization, 
in favor of the like pollen, takes place in Upland, as well as in Egyptian 
cotton. The purpose of this paper is to describe the experiments which 
yielded this evidence. 
1 Accepted for publication Dec. 14, 1923. 
* Reference is made by number (italic) to “Literature cited,*' p. 340. 
Journal of Agricultural Research. Vol. XXVII, No. 6 
Washington. D. C. Reb. 9,1924 
Key No. G-359 
74026—24-1 
(329) 
