PROLIFERATION OF INTERNAL TISSUES OF BUDS. 47 



nearly mature to put forth new growth. If the presence of the larva 

 at this stage is sufficient to cause the bud to fall off, the development 

 of the parasite to maturity is well assured, the large bud affording 

 good protection and adequate food. 



In the Kekchi cotton, however, such late attacks very seldom cause 

 the bud to fall off. Larva? developed in the larger buds are turned 

 out of doors, as it were, by the opening of the flower. The tendency 

 of injured buds to persist is notably greater than in the United 

 States, either because of some physiological difference between the 

 varieties, or because of the larger and more firmly closed involucres 

 of the Kekchi cotton, which keep the buds surrounded with a moist 

 atmosphere and protect it against drying out while the new tissues 

 are forming to heal the wound and encyst the egg. 



In the closely planted Indian fields the squares seldom flare as in 

 the Texas varieties. They generally remain in place and continue to 

 grow until the bracts have reached nearly their full normal size. In 

 fields partially protected by the keleps the weevil larva? do not 

 seem to develop in buds as small as in Texas. Proliferation may 

 partly explain this delay and also the more firmly closed involucres, 

 but in our unprotected plot the weevils were able by repeated punc- 

 tures to infest smaller squares and reach maturity in them, after they 

 had fallen to the ground. 



The behavior of weevil larva? inside the squares in Guatemala 

 seems also to differ appreciably from that observed in Texas where 

 younger squares are usually much more accessible to the weevils, and 

 are commonly punctured. In Texas the larva? regularly grow to 

 maturity, depending for food upon the pollen, which is completely 

 eaten out. In Guatemala this very seldom occurs. Small square^ 

 with well-developed weevil larva' are rarely found under normal con- 

 ditions, nor do the larva? depend upon the pollen as their principal 

 article of diet, as in Texas. 



Several reasons for this difference may be considered. The first is 

 that the larger and more firmly closed involucre of the Kekchi cotton 

 gives the buds several days of protection, so that the average size 

 would naturally be larger. The examination of large numbers of 

 squares picked at random from the Indian cotton fields by Messrs. 

 Kinsler and McLachlan show also that a very large proportion of the 

 punctures are followed by proliferation, and that this means of pro- 

 tection is much more efficient in the younger squares. Another rea- 

 son must be sought, however, for the failure of the larva? to eat the 

 pollen of the large buds where proliferation is less prompt and less 

 frequent. The impression might be gained that the pollen of the 

 Kekchi cotton is in some way not acceptable to the weevils, since even 

 when there is an abundance of pollen at hand they prefer to eat out 



