NATURAL DISSEMINATION. 87 



their exhausted condition. Therefore, the line of continuous infesta- 

 tion may be considerably behind the line of continuous infestation 

 resulting from the last movement of the preceding year. Outside of 

 this continuous line is a strip of considerable width in which the 

 weevil is found scatteringly. Thesummer nights cause these isolated 

 infestations to coalesce. 



FALL DISPERSION. 



All movement of the weevil at other seasons is insignificant in 

 comparison with the great dispersion of the fall which carries the 

 insect far into new territory. It is this movement which causes the 

 more or less regular annual advance in the cotton belt. In one 

 sense this dispersion is merely an overflow from territory in which 

 the insects have become so numerous that there remain no oppor- 

 tunities for breeding. In another sense it appears to be the result 

 of a strong instinct which the weevils possess to invade new regions. 

 At any rate, they show great activity in the late summer and fall. 

 The main causes of the fall flight, therefore, appear to be (1) a 

 scarcity of food and breeding places due to maximum infestation, 

 and (2) an instinct to invade new territory. Several conditions 

 may tend to precipitate the movement or strengthen it. Among 

 these are damage by other cotton insects, which hastens maximum 

 infestation, and drought, which may have the same effect by pre- 

 venting the continued fruiting of the plants. 



There seems to be no special tendency to fly in any particular 

 direction, although prevailing winds frequently cause the majority of 

 the insects to follow one course. This has been observed to be 

 southeast, north, and east in different localities. If not governed 

 by the wind, any weevil wdrich takes flight is as likely to fly toward 

 the old infested territory as in any other direction. It is, therefore, 

 only a portion of the dispersing weevils which enlarges the infested 

 territory. 



The distance any weevil will fly in this movement depends upon 

 how soon it finds uninfested cotton. If on the first flight it finds only 

 heavily infested cotton or none at all it will take wing again. In this 

 way a succession of flights may carry the insect over a wide territory. 

 In one case a distance of over 40 miles has been known to be cov- 

 ered in this manner. If, on the other hand, the first flight carries 

 the weevil into an uninfested field it remains there. Consequently, 

 the advance is slowest in regions where cotton fields are numerous. 

 The occurrence of the leaf w T orm, Alabama argillacea, in great numbers 

 in any locality destroys the food and tends to cause decidedly longer 

 flights of the dispersing weevils. 



So far as we have been able to discover, the weevil has no sense by 

 which it can locate cotton. Such a sense may exist, but the general 

 aimless flight of thousands upon thousands of individuals seems 

 sufficient to account for the infestation of all fields in new territory. 

 An interesting observation w T as made by the junior author and Mr. 

 G. N. Wolcott near Meridian, Miss., that the early dispersing 

 weevils, in flying through hill country with heavy woods, found 

 only the patches on the tops of the hills and from these gradually 

 spread downward to the denser cotton. 



