REPORT OF THE ENTOMOLOGIST. 237 



unobserved tumbling of the larva from higher levels, but where the in- 

 sects have been observed to issue from the bottoms of cellars 10 feet 

 deej), the information would certainly seem to bo reliable. 



METHOD OP BURROWING OF THE LARVA. 



''The method of burrowing and of making its cells is quite interesting. 

 It, scratches away the walls of its cell with the tarsal claw just as one 

 woukl do with a pick, and if it is rising so that the earth removed natu- 

 iall3' falls to the posterior end of the burrow, it simply presses the dc- 

 taclied portions on all sides, and especially on the end of the cavity, by 

 means of its abdomen and middle and hind legs. If, however, it is 

 burrowing downward and the loose soil has to brf pressed against the 

 tip of the cavity, it uses its broad front femora very dexterously in 

 making a little pellet of the soil and in placing it on the clypeal or 

 front part of the head, when the load is carried up and pressed against 

 the top of the cavity. 



" The motions made in cleaning its forearms remind one very forcibly 

 of those made by a cat in cleaning its face. The femora and bent tibioe 

 are rubbed over the clypeus, the numerous stiff hairs on which act like 

 a comb or brush in freeing the si>ines of dirt. 



THE TRANSFORMATIONS. 



[Plate I.] 



"As the time approaches for the issuing of the pupa, it gradually rises 

 nearer and nearer to the surface, and, for a year or two before the ap- 

 pearance of any given brood, this pupa may be dug up within one or 

 two feet of the surface. 



" In the year of their ascent, from the time the frost leaves the ground, 

 they are found quite close to the surface, and under logs and stones, 

 seeming to await the opportune moment, and apparently without feed- 

 ing. They begin to rise from about the 20th of May in more southern 

 localities, and but little later further north. Here, in \Yashingtou, the 

 present year, they begun to sparsely issue about the 23d, and were, per- 

 haps, most numerously rising on the night of the 27th. Those in tlic 

 city were somewhat earlier than those in the woods just over on the Vir- 

 ginia side. The unanimity with which all those which rise within a cer- 

 tain radius of a given tree crawl in a bee-line to the trunk of that tree 

 is most interesting. To witness these pupa? in such vast numbers that 

 one cannot step on the ground without crushing several, swarming out 

 of their subterranean holes and scrambling over the gronnd, all con- 

 verging to the one central point, and then in a steady stream clamber- 

 ing uj^ the trunk and diverging again on the branches, is an experience 

 not readily forgotten, and affording good food for speculation on the 

 nature of instinct. The i^henomeuou is most satisfactorily witnessed 

 where there is a solitary or isolated tree. 



" The pupre (PI. I, Figs. 1, 2) begin to rise as soon as the sun is hidden 

 behind the horizon, and they continue until, by 9 o'clock, the bulk of 

 them have risen. A few stragglers continue until midnight. They in- 

 stinctively crawl along the horizontal branches after they have ascended 

 the trunk, and fasten themselves in any position, but preferably in a 

 horizontal position on the leaves and twigs of the lowermost branches. 

 In al)out au hour after rising and settling, the skin splits down the mid- 

 dle of the thorax from the base of the clypeus to the base of the meta- 



