EEPOET OP THE ENTOMOLOGIST. 267 



terior part of its body to one side, and then , rapidly moving it to the other, 

 lets go at the proper moment ; the momentum of the anterior part of the 

 body is sufficient to carry the whole body some diBtance. In this way a 

 larva can jump two feet in a horizontal direetion. They vrill often spring 

 li-om the highest part of the cotton plant and fall to the ground. On 

 one occasion (August 26) I was in a tiold where the plants v/ere nearly 

 stripped of their leaves at the top ; the larvae were moving to the lower 

 leaves. I saw none crawling down the stalks. All, so far as observed, 

 performed the journey by jumping. They rarely fail to alight upon tbeir 

 feet and cling to the object touched. Kot one in fifty strikes one leaf 

 an<l falls to another before getting hold with the hooks with which the 

 prolegs are furnished. Many, springing too far from the plant, would 

 touch no leaf and thus fall to the ground. I have been unable to find 

 any sdk connecting the larva with the object from which it springs ; and 

 I am of the opinion that in jumping it does not spin a thread. 



I did not observe a single instance of systematic marching, as is indi- 

 cated by the popular name army-worm, which has been so generally 

 applied to this species. I saw on several occasions immense numbers 

 of the larvae on the ground, crawling in all directions in search of food 

 or places in which to transform. And on one occasion I saw mjTiads 

 of the worms of different sizes crawling in all directions over the ground, 

 when there v/as plenty of food and x>laces in which to transform on the 

 plants, as not more than one-third of the foliage had been* eaten. This 

 was the time when I observed so many larvae springing from the stripped 

 upper portions of the plant to the leaves below ; perhaps most of the 

 worms on the ground were those which, in jumping, had failed to alight 

 on the lower leaves. I visited the field at night to ascertain if the march- 

 ing was kept up at that time. I found none crawling over the ground, 

 and nearly all those on the plants were perfectly at rest. 



When the larvae are feeding on the cotton in great numbers there 

 arises a x>eculiar sweetish odor, which, although not easy to describe, 

 is very characteristic. This odor, I supposed, proceeded from the excre^ 

 ment of the larvae ; but Mr. Trelease is of the opinion that it is " due 

 partially to the crushing of the leaves by so many mandibles." In any 

 case this odor is perceptible orjly when the larvae are present in great 

 numbers. The fact that many planters say that they can smell the 

 worms sooner than they can find them otherwise is very strong evidence 

 of the lack of proper knowledge of the habits of this species. 



Although, as a rule, the cotton-worm feeds only on the leaves of the 

 cotton plant, it is occasionally found lying witliin the open flowers feed- 

 ing upon the stamens. It also frequently destroys the buds and small 

 bolls. This is the case when the plant is stripped of its foliage. I have 

 also seen many buds and bolls destroyed when the foliage on the lower 

 thhd of the plants was eaten but little. When a cotton- worm destroys 

 a boll, it does not, like the boll-worm, merely eat out its contents, but 

 often eats the greater part of the pod also. 



From what has been learned respecting the time required for the full 

 development of the larva, and the small amount of injury done during 

 its early stages, it can be seen that the accounts which are often heard 

 respecting the short time which elapses from the first appearance of the 

 worms to the complete destruction of the crop are founded on an error. 

 Wo have heard many accounts of instances where fields had been 

 attacked by cotton-worms and destroyed within three days! If by 

 *' first appearance" one understands the earliest time at which a brood 

 of cotton-worms has been developed of sufiicient size, both as to indi- 

 viduals and numbers, to be easily seen, these accounts will not convey 



