34 



NATURAL CONTROL OF THE COTTON BOLL AVEEVTL. 



For this comparison the figures given in Table III are combined 

 for bolls and for squares. ' 



Table V. — Mortality in bolls versus squares. 



Bolls.... 

 Squares. 



Total 

 forms ex- 

 amined. 



Total 

 weevil 

 stages 

 found. 



Per cent 



of stages 



alive. 



Per cent of mortality 



caused bj'— 1 



Heat. 



Ants. 



Para- 

 sites. 



40,044 

 46,912 



10,051 

 28,940 



64.5 

 36.1 



9.5 

 24.5 



18.4 

 30.0 



3.6 



4.6 



Per cent of 



total 

 mortality, 



three 

 factors. 



31.6 

 59.1 



Several general conclusions may be drawn from the study of these 

 figures. In bolls nearly two-thirds of all the stages found were alive, 

 while in squares the living stages constituted but slightl}^ more than 

 one-third of the total stages found. Pleat is about two and two- 

 thirds times as effective against the weevil stages in squares as against 

 those in bolls. Ants are also much more effective in raising the 

 mortality in squares. Parasites are slightly more active in attacking 

 weevil stages in squares than in bolls, but the difference in this case 

 is much less than in the case of ants or heat. From the combined 

 effect of these three factors there is a total mortality averaging in 

 bolls a little less than one-third, and in squares nearly three-fifths, 

 of all weevil stages present. 



Among the 38,991 weevil'stages found, 63.5 per cent of this total 

 number occurred in fallen squares. It is evident, therefore, that the 

 factors of natural control which are most effective in destroying 

 weevil stages in fallen squares are the most important in restricting 

 the summer multiplication of the weevil. It appears that the Solen- 

 opsis ants and heat are the two most important factors in this con- 

 nection. 



In these examinations it was impossible to account for eggs and 

 ver}" small larvae, as the dr3nng of much of the material and the decay 

 present in another large portion would inevitably efl'aco all traces of 

 these stages even in cases where the work of the living larva^ was not 

 sufiicient to have done so. In the absence of positive data showing 

 the mortality in these young stages, no attempt will he made to esti- 

 mate the mortality that undoubtedly does occur early in the weevil's 

 development. It is certain that the average of the smallest larva^ 

 found would not be more than one-third or possibly one-fourth 

 grown, but the data given for the stages above the one-third grown 

 larva are thoroughly reliable. 



It is therefore " safe to say that the mortalit}^ of the boll weevil 

 throughout the infested area, as re])resented by twenty-eight well- 

 distributed localities, during the period from June 16 to October 15, 

 1906, from the effects of heat or drying, ant, and parasite attack 

 averaged not less than 65 or 70 per cent of all the weevil stages sur- 



