BUREAU OF PLANT INDUSTRY. 267 



in Arizona in dealing with this problem. Breeding work with the 

 Pima variety has been continued, the principal objective being to 

 produce a strain yielding more than the present commercial stock. 

 Encouraging progress has been made in this direction, as well as in 

 tne development of a smoother seeded strain better adapted to gin- 

 ning on the roller gin. 



Investigation of the phenomena of pollination in cotton since 

 the publication of Department Bulletin 1134, entitled " Self-Fertili- 

 zation and Cross-Fertilization in Pima Cotton," has established the 

 fact that pronounced selective fertilization occurs in the upland as 

 well as in the Egyptian type; in other words, when pollen of both 

 types is present in equal quantity on the stigmas of the same flower, 

 most of the ovules are fertilized by the like pollen rather than by 

 the foreign pollen. This fact helps to explain the comparatively 

 low percentages of hybrids produced when Egyptian and upland 

 cottons are grown in close proximity and exposed to natural cross- 

 pollination. 



Hybrids between two strains of Pima cotton, one of which shows 

 a normal development of the conspicuous red spot at the base of the 

 petal and the other complete absence of the spot or its presence as a 

 mere trace, have been studied intensively, and convincing evidence 

 has been obtained that the "spotless" condition is a simple Men- 

 delian recessive. While the results are chiefly of scientific interest, 

 they indicate that such a character as " spotless " flowers could be 

 transferred readily to a commercially desirable strain and might be 

 useful as a " hall mark " in roguing to keep the seed pure. 



Investigations are in progress to determine the degree of isolation 

 which should be given a seed-increase field in order to afford reason- 

 able assurance that accidental cross-pollination will not occur under 

 Salt River Valley conditions. 



BINDER-TWINE FIBERS. 



With the present shortage in the production of Yucatan henequen 

 and the prospect that this shortage will continue for a number of 

 years, the problem of providing for an adequate future supply of 

 binder-twine fiber has been given especial attention. The only per- 

 manently satisfactory solution of our binder-twine fiber problem is 

 to increase the production of this fiber in the insular possessions 

 of the United States. While Porto Rico and the Virgin Islands 

 offer possibilities for a limited production of henequen and sisal, 

 the principal source of supply of binder-twine fiber other than that 

 obtained from foreign countries will be the Philippine Islands. 



A few years ago cooperative work was begun with the Philippine 

 Bureau of Agriculture to encourage the increased production of 

 sisal and maguey fiber in the Philippine Islands. In view of the 

 rapidly increasing consumption of abaca (manila hemp) for binder- 

 twine purposes, this cooperative work has been expanded to include 

 necessary work with abaca. It is entirely possible, if not probable, 

 that the ultimate solution of our binder-twine fiber problem will be 

 an increasing substitution of abaca for henequen in the manufacture 

 of binder twine. 



During the past year large quantities of abaca have been used 

 for the manufacture of binder twine, and it has been demonstrated 



