HIBERNATION. 99 



with whom we have conversed on the subject informs us that in each 

 locality the worms first appear on a certain plantation, and on a very 

 limited part of that plantation. We examined several of these places 

 carefully, bat found no striking local peculiarities. They all agree, how- 

 ever, in being on low land and where the cotton has a thrifty growth. 

 In connection with this testimony of the planters, we must take into ac- 

 count the fact that they seldom observe the worms till the latter part of 

 June or even till July. It is evident, therefore, that it is the first " crop " 

 of worms that appear in the above described localities, and that the 

 testimony has but little bearing ou the origin of the first brood. As yet 

 we have but little data upon this point ; but that which we have indi- 

 cates that the first brood of worms is scattered indiscriminately over 

 those sections in which they occur. Specimens of the first brood were 

 found by Mr. Trelease on cotton growing on bottom land, in a swamp, 

 on an elevation rising from this, and on a ridge considerably distant from 

 the swamp. Thus no local peculiarities of the soil seem to influence the 

 distribution of the worms, except that where the cotton is the earliest 

 the moths first find a place to oviposit. 



We have therefore a very interesting problem presented to us. Why 

 is it that if individuals of the first brood of worms occur indiscrimin- 

 ately on cotton growing on wet and on dry land, that the greater pro- 

 portion of the second or third brood (the first crop) is found only on 

 low, wet lands? The only explanation we can offer, so far, is that in the 

 wet lands there is but little to check the natural increase of the species ; 

 while in dry lands the predaceous insects, especially ants, destroy a 

 large proportion of the larvae of the earlier broods. This point will be 

 referred to again in the chapter on influence of weather. 



It has often been asserted, especially by those who advocate the 

 theory of immigrations of the moth, that the cotton- worm appears first 

 in the western and southern portions of the cotton belt, and progresses 

 regularly toward the east and north. But this does not seem to be the 

 case. As we have already shown, in the spring of the present year 

 (1879) the worms were in Central Alabama as early as there was food for 

 them. And in 1873, when the first brood was so large as to attract gen- 

 eral attention, the worms appeared simultaneously (during May) in 

 Jackson County and Gadsden County, Florida, Decatur County, Geor- 

 gia, Marion County, Mississippi, and Atascosa County and Victoria 

 County, Texas.* 



HIBERNATION. 



How does the cotton- worm pass the winter ? This is a question most 

 often asked respecting this insect, and as yet the answers have been 

 only theories. Many have believed that the pupae of the last brood pass 

 the winter in the ground. This we now know cannot be the case, as the 

 larvae of the last brood web up in leaves in a similar way as do the 



* Monthly reports of the Department of Agriculture, 1873, p. 23-9. 



