122 THE MEXICAN COTTON-BOLL WEEVIL. 



The number of generations decreases with the mean temperature and 

 mean humidity. 



Hibernation seems to begin at mean temperatures between 56° F. 

 and 60° F., but is hastened by high humidity. Cold nights followed 

 by warm, still days seem to stimulate the weevils to considerable 

 activity in the fall, evidently warning them to seek hibernation quar- 

 ters. The period of entrance into hibernation is much more rapid as 

 the mean humidity and mean temperature become lower. The 

 emergence of the weevil is in like manner influenced by the tempera- 

 ture, but it must be considered that the actual temperature experi- 

 enced by the weevil is that which affects the emergence. The time of 

 emergence apparently depends upon an accumulation of a certain 

 amount of effective temperature and a certain amount of rainfall, but 

 if more than the necessary temperature accumulates less rainfall will 

 be needed, and vice versa. The majority of weevils emerge at mean 

 temperatures between 64° F. and 78° F. The percentage of survival 

 seems to decrease as the absolute minimum temperature decreases 

 and the rainfall increases. 



The foregoing statements are conclusions based in some cases upon 

 more or less fragmentary information, but in other cases they may 

 almost be considered as laws of climatic control. 



FIELD OBSERVATIONS ON MORTALITY DUE TO HEAT AND DRYNESS. 



Heat and dryness affect the weevil in a very simple manner. Un- 

 less the square remains moist the food supply becomes unsuitable. 

 In other cases the heat itself causes death directly. Therefore, the 

 hotter and drier the ground upon which the infested square falls, the 

 more certain is the death of the weevil. 



In the years 1906 to 1909 an exhaustive study was made of the 

 effects of various climatic and other agencies which control the boll 

 weevil. In this work 222,715 cotton forms (including bolls and 

 squares) were collected by agents of the Bureau of Entomology at 65 

 localities in Texas, 26 in Louisiana, 7 in Oklahoma, 6 in Arkansas, and 

 6 in Mississippi. Careful laboratory observations were made to deter- 

 mine the mortality due to heat or dryness and to other factors. 



By reference to the series of general tables (LV-LVIII) at the 

 beginning of the discussion of natural control it will be noticed that 

 climatic control kills practically one-fourth of the developing stages, 

 the average for the four years in which records were made being 24.56 

 per cent, which was slightly surpassed by the total insect control. 

 The highest average climatic control was obtained hi 1907, being 29.06 

 per cent, while in 1908 it averaged only 21.21 per cent. 



In rearranging the data to ascertain the condition in which the 

 control was greatest we find the following results: Fallen squares, 

 27.50 per cent; hanging dry bolls, 19.56 per cent; hanging dry squares, 

 17.67 per cent; and fallen bolls, 16.52 per cent. 



The geographical distribution of climatic control is very interesting. 

 In fallen squares the various sections ranked as follows: North-central 

 Texas, 43.99 per cent; southern Texas, 34.10 per cent; central Texas, 

 31.28 per cent; eastern Texas, 29.31 per cent; northeastern Texas, 

 27.89 per cent; Arkansas, 26.54 per cent; Oklahoma, 22.62 per cent; 

 Mississippi, 15.63 per cent; southwestern Texas, 13.38 per cent; 

 Louisiana, 13.11 per cent. 



