8 BULLETIN 233, U. S. DEPARTMENT OF AGRICULTURE. 



covered, open back flat in a very abnormal position and become 

 pale, sickly yellow. Such an injured square is very conspicuous on 

 a plant in the field and is usually the first indication noted of the 

 presence of weevils. After a few days the square falls to the ground 

 in nearly all varieties of cotton, and in heavily infested fields in the 

 South it is a very common sight to see great numbers of these squares 

 scattered beneath the plants. With the bolls the injury is not so 

 easily noted, since they do not fall unless very heavily infested, but 

 the punctures are readily found by a careful examination, and fre- 

 quently the form of the boll is distorted. (Pis. Ill and IV.) 



FOOD PREFERENCES OF THE ARIZONA WEEVILS. 



A number of tests have been made to determine whether or not 

 the Thurberia weevil displays any preference for either Thurberia 

 or cultivated cotton. These experiments were conducted both in 

 the laboratory and in large cages in the field, and great care was taken 

 to ehminate all factors from the choice other than the actual attrac- 

 tion of the plants. It was found that individuals removed from 

 hibernation cells and offered both plants displayed what seemed to be 

 only the slightest choice in favor of Thurberia, and this disappeared 

 after a few days' feeding. Weevils removed from the cells and fed 

 only upon Thurberia for a few days and then offered a choice at first 

 displayed a marked preference for Thurberia. After a few days' 

 feeding in the presence of both plants this preference gradually dis- 

 appeared and cotton was as much eaten as Thurberia. Weevils fed 

 only upon cotton for a few days after removal from the cells would at 

 first display a choice in favor of cotton, but this disappeared in the 

 same manner. From these experiments, and also from observations 

 made in the field in 1914, it seems safe to conclude that the weevils 

 have very little inherent preference for either plant and that neither 

 plant has the power to attract them away from the other. 



THE TRANSFER TO COTTON. 



The transfer of the weevil to cultivated cotton may be accom- 

 plished in two ways, i. e., by flight or by water. While it is of course 

 impossible to determine the exact extent of the flight of these weevils, 

 either in distance or frequency, all available evidence seems to indi- 

 cate that this means is likely to be of little importance in the pri- 

 mary spread of the weevils. It seems that as long as there is an 

 abundance of food at hand the weevils will fly very little, but in case 

 of food shortage they fly readily. On the other hand, the habits 

 and present distribution of the weevil make the species particularly 

 adapted to dispersion by floods. Most of the Thurberia plants grow 

 either directly in the wash of a canyon or arroyo or where the surface 

 drainage is directly into such a wash. Many of the bolls containing 



