62 THE MEXICAN COTTON-BOLL WEEVIL. 



when the square will be ready to fall. In especially severe cases of 

 feeding injury flaring often results in less than 24 hours. Occasionally 

 the growth of the square overcomes the injury from feeding, and the 

 bracts, after having flared, again close up and the square continues 

 its normal development and forms a perfect boll. When injured by 

 the feeding of a young larva as the direct result of successful oviposi- 

 tion, flaring was found in 193 cases to take place in an average of 

 7 days from the deposition of the egg. (See Pis. V, VI.) 



After an average period of 2.5 days subsequent to flaring the 

 square was found to fall to the ground, although it may sometimes 

 hang by a thread of the bark. The average time from egg deposi- 

 tion to the falling of the square in 539 cases from June to September 

 was found to be about 9.6 days, which is about the middle point of 

 the weevil development. It has been shown in another place (Table 

 XXVII) that the period before the falling of the square has a direct 

 bearing upon the period of the development of the weevil. 



PROBABLE ORIGINAL BREEDING HABIT. 



There is nothing to indicate that the boll weevil has changed its 

 food plant, although it may have done so. It is now confined, as far 

 as we know, to the various species and varieties of the genus Gossy- 

 pium. The boll weevil belongs to a genus of weevils every species of 

 which is confined in its food habits to a single species or genus of food 

 plants. The majority of the species of Anthonomus and perhaps all 

 that belong to the true genus normally breed in buds. It is therefore 

 reasonable to assume that the normal habit of the boll weevil is to 

 breed in the cotton buds or "squares," and that its habit of breeding 

 in the bolls is an adaptation due to the necessity of providing for the 

 great number of weevils which develop in the later part of the season. 

 A study of the length of the development of many species of Antho- 

 nomus leads the authors to believe that the short developmental 

 period in squares is perfectly normal and that the longer period in 

 Dolls is due merely to environmental conditions, as is explained under 

 the subject of development. 



THE EGG. 

 DURATION OF EGG STAGE. 



Concealed as the eggs are beneath several layers of vegetable 

 tissue, it is impossible to examine them to ascertain the exact length 

 of the egg stage without in some degree interfering with the natural- 

 ness of their surroundings. The beginning of the stage is easily 

 obtained by confining female weevils with uninfested squares. By 

 making a large series of observations about the time that the larvse 

 should hatch it is possible to obtain the average length of the egg 

 stage. The extreme range which has been observed in the duration 

 of this stage is from 1 to 17 days, while the average period for the 

 whole number of observations is but 3.7 days. It is possible that the 

 embryo can undergo an even greater retardation without losing its 

 vitality. The period of embryonic development is lengthened by 

 decreases in the temperature and also by lowered atmospheric 

 humidity. Thus it was found that between 79° F. and 81° F. the 



