DESTRTTTTOX OV COIION lOllMS. 71 



seems that the ])resen('e or abs(MU'o of a. wcM^vil sta*i;e lias little, if niiy- 

 t]iin<i:, to do with tlu^ retention hy the plant of a poiMion of ils surplus 

 fruit, hut that tlu* abundance^ of wccxils in |)i()portion to their food 

 supply may he lar<z:(dy resj)onsil)l(' foi- tlu> variations wliidi a|)p('ar in 

 tlie j)ereentage of these dried, hani^in<^ bolls wbicli contain >onie 

 weevil stage. 



In Table XVI B, anioniz; the hanu^in^ dri(Ml scpianvs, nearly .')'.) per 

 cent of these examined eontain(Ml soinc^ wcM'vil statue. In this case 

 tluMH^ \ver(^ found an aA-(*ra«j:e of but 1.01 stai^es per s(piai-(\ The pei-- 

 cenlage eontaininij: weevil stages vari(\s somewhat, as it did in the ease 

 of dried bolls, but the variations are not exactly ])arallel. One 

 reason for this apparcMit lack of agi-e(Mnent may be found in the faet 

 that weevils show eonsiderable j)referenee for s(piares, and therefore^ 

 attack them to a nuieh largtu* extent than they do the small bolls. 

 Among the bolls an average of 24.6 per eent contained some stage of 

 the weevil, while among the dried scpiares there were 5S.8 ])er cent. 



In Table XVI C, fallen bolls, it appears that an average of only 

 about 18 j)er eent contained some weevil stage. In this case there 

 were found 1.106 weevil stages for each boll found to contain any. 

 Examinations in three localities failed to reveal any weevil stage in a 

 total of 380 bolls. Two of these examinations were made at Victoria 

 in June before weevils had become sufficiently abundant to attack 

 bolls to any extent. The third case was at San Antonio, where, as is 

 indicated by observations in the other classes of forms, the infestation 

 was comparatively light. It is probable, ho\vever, that a more 

 extensive examination would have revealed some weevil stages in 

 fallen bolls at San Antonio. 



In Table XVI D, fallen squares, it is shown that slightly over 60 

 per cent contained some weevil stage. In this very large series of 

 examinations there were but 1.05 stages for each square containing 

 any. As would naturally be expected, there is not qiute so wide a 

 range between the extremes in the percentage of total sqmires which 

 were found to contain a weevil stage as in other classes of forms. 

 The importance of fallen squares, as compared with other classes of 

 infested fruit, is shown by the fact that, in the 36,354 fallen s(|uares 

 examined, were found 22,169 weevil stages; whereas, in th(> tO,l()6 

 forms in the other tliree classes for Texas, there were but 1 1 .(HH) 

 weevil stages. While 60.7 per cent of the fallen squares w(M'e found 

 to contain some stage of the weevil, an average of the three otluu- 

 classes of forms showed stages in but 26 per cent. The special signif- 

 icance of these comparisons, from the standpoint of natural control 

 of the weevil, may be appreciated when it is considered that the total 

 mortality in fallen scpiares is nmch greater than in any other class of 

 forms. 



