April, '12] HUNTER: COTTON MOTH OUTBREAK 125 



they seem to indicate that the insect became extremely abundant 

 very early in the season in the northern states of Mexico. Whether 

 this infestation was the result of flights from more southern localities 

 or of the sudden increase of local colonies is beyond determination 

 at this time. 



In the eastern part of the cotton belt the earliest record of the occur- 

 rence of the cotton caterpillar was at Oswego, South Carolina on 

 July fifth. At that time it was exceedingly scarce and in fact did not 

 assume large numbers until the 20th of September. 



From the locality near the Mexican border in Texas the infestation 

 radiated rapidly northward and eastward beginning about the middle 

 of June. The flight of the moth did not seem^to carry it a great 

 distance in a northward direction. It seemed to obtain an al^undance 

 of food after a short flight and was not compelled to make a series of 

 movements. 



By the middle of July defoliation began in Louisiana and Missis- 

 sippi but was not near complete until more than a month later. In 

 North Carolina the earliest occurrence of which we have any record 

 furnished by Mr. Franklin Sherman, Jr., was on August 10th. On 

 September 19th large swarms of the moths appeared at the electric 

 lights in the city of Washington and the invasions seemed to continue 

 until October 29th. These observations were made by Messrs. 

 Schwarz and Pergande who naturally took great interest in it on 

 account of their work on the insect many years before. Mr. Schwarz 

 informs the writer that he does not remember any equally heavy 

 flight of the moths in Washington since about 1882. This northward 

 flight was apparently the result of a development of the fourth or 

 fifth generation of the insect. It is somewhat remarkable that the 

 appearance of the moths in large numbers at lights occurred practically 

 simultaneously at Washington, D. C, Clarksville, Tennessee and 

 Dallas, Texas. This may possibly indicate that there was a heavy 

 inflow of the moths from South America at about this time. 



Our conclusions regarding the origin of the outbreak of 1911 is 

 that it started from two infestations. One, apparently unimportant, 

 in the eastern part of the cotton belt, and the other of much greater 

 importance, which began in northern Mexico. The filling in of the 

 intermediate territory in the Mississippi Valley, however, can hardly 

 be explained satisfactorily on the theory that these two invasions 

 increased and eventually coalesced. In fact the general heavy infesta- 

 tion which became noticeable throughout the south early in September 

 can only explained on the assumption that an invasion of moths from 

 across the Gulf of Mexico took place shortly before that time. 



With the sudden increase in the moths there arose an enormous 



