6 



BULLETIN 926, U. S. DEPARTMENT OF AGRICULTURE. 



The breeding experiments of Coad and Howe, botli published, show 

 that the weevil can live on Hibiscus, and Coad reared an adult from 

 Hibiscus, but the weevil can not normally attack the plant. So far 

 as we know, there are no plants growing in Florida or the southern 

 States that are adapted to serve as food plants of the boll weevil. 



THE BOLL WEEVIL ON SEA-ISLAND COTTON. 



FEEDING EXPERIMENTS TO DETERMINE THE PREFERENCE SHOWN BY THE BOLL 

 WEEVIL FOR SEA-ISLAND COTTON. 



A known number of adult boll weevils were placed in battery 

 jars and allowed to feed on bolls, squares, and leaves. Each jar 

 contained both sea-island and upland bolls, squares, and leaves, and 

 careful records were taken every two hours to determine whether 

 there was any preference by the weevil for either of the two types 

 of cotton fruit and foliage. The experiments were conducted in 

 an outdoor insectary where normal atmospheric conditions prevailed. 

 The data obtained in order to ascertain whether there is any pref- 

 erence by the boll weevil for sea-island as compared with upland 

 cotton are given in Table I. The records presented in this table show 

 that the boll weevil seems to prefer sea-island foliage to upland 

 foliage, and that it shows a distinct preference for sea-island bolls 

 compared with upland bolls. It appears also that there is a tend- 

 ency to feed on upland squares in preference to sea-island squares. 



Table I.- 



-Feeding choice of boll weevil between sea-island and vpland cotton, 

 Madison, Fla. 





Number 

 weevils 

 used in 

 experi- 

 ment. 



Total 

 number 

 indi- 

 vidual 

 obser- 

 vations . 



Total number of 



times weevils were 



observed. 



Percentage of 



feedings recorded 



on — 



Food. 



Feeding 



on 

 upland 

 cotton. 



Feeding 

 on sea- 

 island 

 cotton. 



Resting 



on 



cage. 



Upland 

 cotton. 



Sea- 

 island 

 cotton. 



Bolls 



10 

 10 

 10 



290 

 290 

 290 



64 

 135 



77 



118 

 110 

 129 



108 

 44 



84 



22.0 

 46.9 

 26.5 



40.6 





37.9 





44.4 







Total 



30 



870 



276 



357 



236 



!31.8 



'40.9 







1 Average. 



In Table I attention is called to the preference by the boll weevil 

 for sea-island bolls over upland bolls in the experiments conducted 

 on the feeding habits of the weevil. In 1917, in order to secure fur- 

 ther data on this phase of the weevil's attack, the writer examined at 

 random each week 100 grown bolls of each variety of upland and sea- 

 island cottons that were still green, beginning on the 19th of July, the 



