15 



Further investigations will probably I'esult in the discovery of other 

 races of cowpeas which may be so improved by selection that they 

 may be planted on land infected by the wilt disease. 



The wilt disease of watermelons, also allied to the two preceding, 

 may prove amenable to the same treatment. The Department has 

 under way some experiments to determine the possibility of finding 

 a race of watermelon which may be grown on infected land. This 

 would be exceedingly desirable, for this disease has made the growing 

 of melons for market impossible over large areas in the South which 

 formerly produced them in great abundance. 



CONCLUSIONS. 



There is great promise of a successful i-emedy for the cotton-wilt 

 disease in selection of seeds from healthy plants growing on infected 

 soils and by continuing to select and cross-breed the most resistant 

 plants in succeeding crops with a view both to resistance and quality 

 of staple. 



It would be well in the case of upland cotton to start with a race 

 like the Jackson, which is already highly resistant, and improve and 

 fix the quality by careful cross breeding and selection. In places 

 where this cluster type of cotton is undesirable a resistant strain of 

 the sorts commonly cultivated can probably be obtained by cross 

 breeding and selection. It is hardlj^ to be expected that this process 

 will result in a perfectly immune race the first year. Even though 

 much of the cotton become diseased, the selection should be con- 

 tinued each succeeding year until the quality of resistance is fixed. 



In the case of the sea island cotton, where length and fine quality 

 of staple are essential, the process of selection and breeding should 

 l)e the same. Resistance to disease must be the primary requisite, 

 and from the resistant plants those bearing the finest lint may be 

 selected. , 



The Egyptian cottons will i^robably prove of the greatest value when 

 crossed with our ui)land races so as to add the vigor and quality of 

 the former to the productiveness of the latter. It is hoped that the 

 Depai-tment will be able to extend its work along this promising line. 



In addition to selection for resistance, all practicable ijreventive 

 measures should be applied. Rotation of crops is even more impor- 

 tant on these infected soils than on healthy ones, for the continual 

 growing of cotton on these lands will increase the amount of disease 

 and decrease the resistance of the cotton. 



Prompt destruction of diseased plants is also very important. 

 Every effort should bo made to avoid the infection of healthy fields 

 l)y anijnals, tools, wash watei- from diseased fields, diseased plants, 

 infected compost, etc. As already stated, land once infected with 

 this disease reniains infected lor an unknown period. 



