A CONSIDERATION OF THE CULTURAL SYSTEM. 109 



the other seasons mentioned. It was due to the prevailing tempera- 

 ture conditions. The killing frost was preceded by several light 

 freezes, which, however unpleasant to the weevil, would not induce 

 the insect to enter hibernation, since the days were invariably warm 

 enough for the weevil to continue feeding. The experiment-station 

 cotton was thoroughly frozen on the night of November 27 — two to 

 three weeks later than in the last two years — after which the weevil 

 disappeared from the cotton field in the course of three days. These 

 observations were made on the experiment station's own cotton. 

 During the fall of 1903 observations were made on cotton stalks 

 burned before frost, the effect of which, however, could not have 

 been fully brought out on account of the miscellaneous purposes for 

 which the plat was used the following year. During the season of 

 1904, in addition to the weevils already present in the fall, thousands 

 of specimens were brought in from other fields to be used in experi- 

 mental work, so that there was every opportunity for a large per- 

 centage of hibernating weevils to pass through the winter, a wooded 

 ravine very near the cotton plants offering shelter. During the early 

 days of September the cotton leaf -worms {Alabama aryillacea Hbn.) 

 made their appearance. They increased rapidly^ and about October 

 10 there was not a green particle of cotton left in the field. The great 

 number of weevils present disappeared. No opportunity was given 

 the plants to put forth leaves again during the next rain, but all 

 stalks were destroyed, put up in piles in the field, and burned about 

 a week after the first frost, when it was certain that all weevils were 

 in hibernation. The ravages of the cotton leaf-worm gave us the 

 same condition as though we had destroyed the stalks about October 

 10. It followed that the injury to cotton this season amounted to 

 nothing, notwithstanding the fact that we were delayed in planting 

 in spring. 



The cotton leaf-worm gave us a very conclusive demonstration of 

 the value of the fall destruction of cotton stalks. It the cotton had 

 not been destroyed the preA T ious fall, damage would have been done 

 by the weevil, yet without the thorough cultivation the absence of the 

 weevil would not have secured us such a heavy cotton crop, which was 

 fully shown by observations made on other plantations during the 

 drought. The cotton leaf -worm, which for many years has been a 

 great enemy to the cotton planter in the cotton States, was a very 

 efficient ally during the fall of 190-1 in destroying the hibernating 

 brood of boll weevils on Upland cotton at College Station. During 

 1905 its efficacy in this respect amounted to nothing. The frequency 

 with which it can be depended on varies in different localities. 

 Some planters report that the worm is destructive in their localities 

 during fall nine years out of ten. The writer has given careful atten- 

 tion to the study of the practical side of predaceous and parasitic 



