210 REPOET UNITED STATES ENTOMOLOGICAL COMMISSION. 



of the locust from properly drying after his fourth molt, but a clear, sunny day not 

 only promotes this, but, if accompanied by a dry atmosphere, there is produced a rapid 

 evaporation, a nervous irritating dryness and stiffness, or some other state that stimu- 

 lates the locust to take recourse to flight. Once on the wing, however, he is borne by 

 the prevailing wind. The migration thus initiated by the dryness of the wind is con- 

 trolled by the strength of the wind, while it seems due to the volition of an intelligent 

 animal in search of distant green pastures in the lower lands to the southward. The 

 dry winds of the West are generally the N., NW., and W. 



3. Conversely, the locusts hatched in moist lowlands find their molting stages re- 

 tarded, and the full-fledged insect being uncomfortable with his partially-stiffened 

 limbs and wings, takes to flight as a relief, and is again borne along by the wind ; but 

 it is now a moist wind that is beaming him onward. 



The moist winds in the lower States (Missouri, &c.) are from the SW., S., and SE. ; 

 and as these are also largely prevalent, they produce the appearance of a return mi- 

 gration. Tlie enfeebled condition of these lowland-born insects is apparently made up 

 for by the greater frequency and strength of the winds ; but only a few, and these, of 

 course, the strongest, ever regain the Rocky Mountain breeding-grounds. It is, then-, 

 the relative humidity of the air that determines whether the locust will fly or not. 

 Its nature is adapted to a certain range of atmospheric moisture j air drier or damper 

 than this is uncomfortable to it. 



These hypothetical views find some support in the following table, in which I have 

 given, for each of our three epochs in locust-life, the year during which favorable and 

 unfavorable winds w^ere specially frequent at each locality, and by which, therefore, a 

 judgment can be formed as to whether the years in question were any of them likely 

 to be on the whole favorable to a grasshopper invasion. The conclusions are, of course, 

 qualitative only and not quantitative, on account of the want of numerical data for 

 the relative humidity, &c. 



