THE VAGABOND CRAMBUS : ITS OPERATIONS. 131 



seemed doomed. I counseled a delay of a few days, until some ex- 

 periments could be made of means which might perhaps prevent far- 

 ther depredations. It was suggested that strips of about ten feet in 

 Avidth should be staked off in the meadow, and that some harmless 

 substance (not daring to recommend poisonous ones on meadow land), 

 which might render the grass unpalatable for food to the larvge, be 

 scattered over each strip. Upon one plaster might be used, upon the 

 next lime, upon another ashes, on another salt, and on a fifth road- 

 dust. These applications were to be made when the grass was wet 

 with rain or dew, and renewed if washed away by the rain. The rela- 

 tive effect of the several applications was to be noticed, and if effect- 

 ual in arresting the depredations, that which gave the best result 

 should be employed at large. 



Upon another farm, that of Mr. E. C. Green, I examined a side-hill 

 pasture of several acres, of which about two acres had been destroyed. 

 A line separating this portion from the uninjured part, running down 

 the hill, was distinctly traceable, but no cause for its existence could 

 be discovered, either in the character of the soil or conformation of the 

 ground. Perhaps it marked but a temporary hmit, which may have 

 shown an advance the following day. Another field below, on quite 

 low ground, was so entirely browned that I could not convince those 

 with me that it had not been winter-killed, until by stooping down 

 and parting the dead grass of the preceding year, I was able to point 

 out a thick growth of new grass closely eaten to the surface of the 

 ground. 



Characteristics of the Attack. 



Sevei-al interesting features characterized this attack. It was in most 

 instances observed first on upland and old pastures, — differing in this 

 particular from the invasions of the northern army-worm, Leucania 

 unipuncta, which ordinarily commences on low grass lands, working 

 thence upward. 



Its progress was remarkably rapid. A brown patch first observed in 

 a pasture would rapidly extend its area, and often embrace the entire 

 field in from ten to twelve days. Whether this was the result of the 

 spreading of the larvae from certain points where the eggs had been 

 deposited, or the unequal hatching of the eggs distributed over entire 

 fields, as influenced by various conditions of warmth, moisture, etc., I 

 could not ascertain, for the brief time which I was able to devote to 

 my observations did not permit of the detection of the eggs or their 

 empty shells. 



The secrecy with which the depredations were conducted was un- 

 usual and surprising. The extension of the infested areas could be ob- 

 served from day to day, seemingly as causeless as a blight. Only a few 



