44 NATURAL CONTROL OF THE COTTON BOLL WEEVIL. 



possible that the indiscriminate work of ants may explain, in part 

 at least, the fact that fallen forms have invariably yielded a smaller ' 

 percentage of parasites than have hanging forms. Referring to 

 Table III, this difference in the percentage of parasite attack is sho^^^l 

 as follows: In hanging bolls, 6 per cent; in hanging squares, 12.8 

 per cent; in fallen bolls, 1 per cent; in fallen squares, 3.2 per cent. 

 As the ants are already distributed in the cotton fields throughout the 

 wee\al-infested area, it is evident that their relative abundance in 

 different localities may explain to a considerable extent the variations 

 wliich have been foimd in the percentage of mortality resulting from 

 ant attack. 



It has been found that all of the parasites which attack the weevil 

 are native species having some other host which lives normally in some 

 part of the vegetation surrounding the cotton fields. In no case can 

 the weevil be considered as the primary or preferred host of any of 

 these species. It has been found that in many instances the occur- 

 rence of other weevils, living under such diverse conditions as in the 

 buds, flowers, fruit, seeds, stems, roots, or in galls formed by the attack 

 of some other insect, a large variety of weeds, or other common plants 

 and of trees, directly influences the abundance of parasite species and 

 explains variations which have been found in the extent of parasite 

 attack upon the boll weevil. It is plain, then, that the abundant 

 occurrence of some plant supporting a large number of other weevils 

 which do not at all affect cotton may produce indirectly an abundance 

 of some parasite species which is capable of attacking the weevil. 

 It is to be hoped that some of these parasites ma^^ ultimately develop 

 a preference for the boll weevil as their host. This would seem to bo 

 a possibility^, because the boll weevil presents a continuous series of 

 stages for parasite attack throughout the season, whereas in many 

 cases it is known that parasites are forced by the short breeding sea- 

 son of their original host to change from one host to another in order 

 to continue their o\\ti reproduction throughout the season. The 

 tendenc}^ of parasites to adapt themselves to this change in possible 

 host conditions demands continued careful investigation. 



The note records concerning cultural conditions in the various fields 

 investigated are not as complete as is desirable for this study, but a 

 general summary of conditions as noted ^dll be given to determine 

 the possibility that cultural conditions may explain some of the 

 variations which have been found in mortality. In Table VIII, Group 

 A, the cultural notes are given for most localities recorded in Table 

 VII, Group A. The same arrangement in regard to total mortality is 

 retained. 



