42 NATURAL CONTROL OF THE COTTON BOLL WEEVIL. 



From the eleven localities in Group A 52,334 forms were examined. 

 This number constitutes 60.2 per cent of the total of the three groups. 

 In these forms were found 23,947 weevil stages. This number consti- 

 tutes 61.4 per cent of the total number of wee\nl stages found in the 

 three groups. The proportion of weevil stages to forms examined in 

 each of the groups varies but slightly. When we come to the total 

 number of weevil stages killed by the three factors, we find in Group 

 A 15,145 dead, which number is 74.7 per cent of the total mortality 

 occurring in the three groups. This shows that in Group A there was 

 actually an increase of nearly 15 per cent in the total mortality above 

 the average proportion which might have been expected in tliis group. 

 The average mot-tali ty occurring in Group A is 63.2 per cent. The 

 percentage of mortality in Group A from each of tlie three factors is 

 somewhat greater than the average sho\Mi in Table VI. For heat and 

 dr^-ing this increase amounts to 6.3 per cent, for ants to 4.5 per cent, 

 for parasites to 0.5 per cent. The greatest increase is tliereforc 

 attributable to heat or drying and it might naturally be expected that 

 the climatic records would show a considerabh^ higher maximum 

 temperature to account for this increased mortality. A comparison 

 of the mean maximum temperature records for Group A with those for 

 Groups B and C shows, however, an average difference of less than '2 

 degrees. Between A and C there is an average difference of but 0.1 

 degree. In group A 27 per cent of the weevil stages found were' 

 destroyed by heat or drying, while in Group C, having almost identic- 

 ally the same mean maximum temperature, the mortality from tliis 

 cause is but 7.1 per cent — a decrease in Group C of 19.9 per cent. 



The percentage of mortality from ants shows an even greater dif- 

 ference between Groups A and C than has just been sho\\-n for heat, 

 amounting in this case to a decrease of 22.8 per cent. Certainly tliis 

 difference can not be attributed to the variations in temperature 

 sho^^^l for these groups. 



^^liile unquestionably an unusual deficiency of rainfall is very 

 important in checking the development of weevils, the difference 

 shown in this table for the average monthly rainfall in each of the 

 three sections can not account for the difference found in the total 

 mortalit}^ from any of the three factors. Thus in Group A, having a 

 monthly average of 3.39 inches, which is 0.89 of an inch above the 

 average normal rainfall for the localities given, and an average of 6.2 

 rainy days per month, the mortality from heat or dr^-ing averages 27 

 per cent. In Group B, for seven localities the average total rainfall is 

 2.6 inches per month. For these localities this is a deficiency amount- 

 ing to 0.72 inches per month. AVe have here five rainy days per 

 month and the mortality from heat or dr^^ng is but 13 per cent, or less 

 than one-half the mortality found in Group A. Similarly, in Group 

 C, with a mean maximum temperature approximately the same as 



