FERTILIZATION IN PIMA COTTON. 43 



RESULTS OF OTHER INVESTIGATORS. 



Balls (7, pp. 222, 223; 8, pp. 122-125), using " a method of mixed 

 pollination, whereby the stigma of a flower received equal quantities 

 of (1) self pollen from its own anthers and (2) pollen from an- 

 other plant," found that the seed produced by Egyptian flowers re- 

 ceiving both self pollen and upland pollen yielded 10 hybrids out 

 of 330 plants, or somewhat less than 3 per cent. The percentage was 

 about the same with the reciprocal cross-pollination upland X self + 

 Egyptian. On the other hand, when stigmas of Egyptian or of 

 upland cotton were pollinated simultaneously with approximately 

 equal quantities of self pollen and of pollen from Egyptian X upland 

 F a plants, the resulting percentages of hybrids were 20 and 28, re- 

 spectively. These results seem to indicate that pollen of the conju- 

 gate generation of a hybrid between very different types of cotton, 

 when applied to the stigmas of one of the parent types, is better able 

 to compete with the self pollen than is the pollen of the other 

 parental type. Balls does not describe in detail the method used in 

 this experiment ; but if automatic self-pollination of the base of the 

 stigmas was not prevented, this would account in part for the very 

 low percentages of hybrids when pollen of the other type, in addi- 

 tion to self pollen, was applied. 



Longfield Smith (44) performed experiments the object of 

 which was to produce the largest possible number of ¥ t hybrids 

 with the least expenditure of labor. To this end the stigmas of 

 unemasculated flowers of sea-island cotton and of cotton of an up- 

 land type, said to be native in St. Croix, were smeared at 7.30 to 8 

 a. m. with pollen of the other species. Sea island X St. Croix yielded 

 from 30 to 40 per cent of hybrids, while the reciprocal cross-pollina- 

 tion yielded 70 per cent. The much greater percentage of hybrids 

 obtained from the reciprocal was attributed by this experimenter to 

 the earlier opening of the anthers in the sea-island than in the St. 

 Croix flowers, which allowed automatic self-pollination to begin 

 earlier in the former than in the latter. The very high percentages 

 of hybrids obtained from unemasculated flowers of the St. Croix 

 cotton in this experiment seem to indicate a decided " prepotency " 

 of the foreign pollen. 



The hitherto unpublished data of an experiment conducted by 

 Argyle McLachlan, under the direction of O. F. Cook, at Yuma, 

 Ariz., in 1910 and 1911 are also of interest in this connection. Flow- 

 ers of Egyptian cotton (Yuma variety) were pollinated with pollen 

 of the same variety and with upland pollen, and flowers of upland 

 cotton (Triumph and Durango varieties) were pollinated with pollen 

 of the same variety and with Egyptian pollen. The flowers to be 

 pollinated were emasculated before their anthers had opened and 

 were then inclosed in bags. The flowers which were to furnish the 

 pollen were bagged before their corollas had opened. Pollination 

 was done as soon as the anthers opened. 26 In some cases the two 

 kinds of pollen were applied simultaneously, in other cases the sec- 

 ond kind was applied after intervals of 15, 30, and 60 minutes, using 



26 The original records of this experiment apparently have been lost, but the results are 

 stated in a memorandum prepared by Mr. McLachlan on August 9, 1911, from which 

 Table 25 has been compiled. Further details in regard to the procedure followed were 

 supplied from memory by Mr. McLachlan in a letter to the writer dated December 

 5, 1919. 





