52 



In such numbers were the larVcT descending the trunk that for sev- 

 eral succeeding days they were repeatedly swept down with a broom 

 before spraying them with kerosene, pouring hot water over them, or 

 sweeping them into piles of leaves for burning. The largest proportion 

 of tlie larvffi, instead of transforming at the base of the tree, crossed the 

 sidewalk and collected against the side of the house and a brick wall 

 adjacent, or in the crevices at the ends and sides of the adjoining first 

 course of bricks in the walk. Many were observed in the attempt to 

 climb up the side of the house and wall, but they almost invariably fell 

 before attaining a height of 2 feet. At one point, where a small shrub 

 of a few inches in height had sprung up against the house, there were 

 swept up on August 23, from an area of 8 by 10 incites, nearly a pint 

 measure full of the larvae and pupoe, with a few of the newly trans- 

 formed beetles. From a partial count it was computed that this collec- 

 tion embraced 9,750 individuals. 



Transformations. — A few facts relating to the transformations of this 

 brood may be of value. The time intervening between the descent of 

 the larv?e and their pupation was not noted ; it probably did not exceed 

 five days, for of 100 taken August 15 almost one half pux^ated on the 

 18th and the remainder on the 19th. Those taken in their curled form 

 l^upated within two days, as observed in 150 examples taken from the 

 base of a tree on the IGth, 5S of which became pupie within thirty hours, 

 and all within forty-eight hours. 



The time from the pupa to the imago was seven days, as was observed 

 in several lots that were separated soon after puliation, which in each 

 instance gave the beetle — at first of a pale tint of yellow, with entire 

 absence of black on the elytra — on the seventh day. 



The descent of the larvae was apparently completed by the 25th, none 

 being seen upon the trunks after this date.* The degree of develop- 

 ment of the brood at this time was about that of the first brood when 

 observed by me on July 15. From this we may infer a period of about 

 six weeks for transformations of the brood, or about a week less than 

 that which has been recorded for the earlier brood in Kew Jersey. 



Do many larvcv drop from the branches f — Writers upon the habits of 

 this insect have stated that many of the larvae drop from the tree 

 instead of traveling down the trunk, particularly when the elms have 

 drooping branches. This would seem reasonable, but it was not con- 

 firmed by my observations. The Scotch elm on my premises which has 

 been referred to has long and somewhat drooping branches, which extend 

 over an extension to the house; bat I have not been able to find at any 



*" There seem to hare been some later than this, perhaps by dropping from the 

 leaves, for they continued to be found in small numbers near the wall until Septem- 

 ber 9. Pupte were noticed until the 12th, and the beetle was abroad until September 

 24. On the 3d of October one came to light in my office in the capitol, of a remark- 

 ably dull color, suggesting the idea that it may have been drawn from its hiding 

 place after its development from the last larvae brought by me within doors. 



