NYMPIIALINAE: BASILARCIHA ARTIIKMIS. 301 



the leaf, and there eats on hoth sitles of tlie nii(h-ih, nsuallv restinjr on the 

 projecting end of this" as on a perch. Jt undergoes one or two moults 

 before winter, during all this time resting, when not feeding, upon this perch 

 or a similar one constructed from the next leaf it attacks ; it then pro- 

 ceeds, after the fashion of Basilarchi.ans, to consti-uct a winter abode, f^en- 

 erally from a leaf which it has been eating. These hil)crnacula are in all 

 respects similar to those made by the other species of the genus; it 

 l)egins to make them in the Wiiite Mountains, and e\cii fiu-ther south, bv 

 the middle of August, and it is sometimes as many as ten days in making 

 one quite to its liking. Into such a Httle cylinder the caterpillar crawls 

 head foremost, its tuberculate hinder end visil)le and forming a sort 

 of living door to its abode : here it remains the winter through, and as it 

 has lived up to this time on the same low plant on which it was born, it is 

 buried in snow the greater and colder part of the winter. This I found by 

 examining thousands of trees by the roadsides in tlie White Mountain 

 valleys very early in spring before the leaves were fairly out, when search 

 for the hibernacula wns easiest, and I found that fully nine out often were 

 within from two to three feet from tlie o-round. much below the averaire 

 level of the snow in that region. With the first starting of the leaves 

 into new life the caterpillar backs out of its retreat and begins again to 

 feed upon the tender green ; but either it first changes its skin, or it 

 returns to the perch of its dried up and inedi!)lc hibei-naculum-leaf after its 

 [)robably scanty meals, for these will nearly always be found, soon after the 

 catei'pillar has finally quit them, to have uj)on the apical Hap, the cast-off 

 pellicle of the caterpillar. 



It may be added as a very interesting point, that there is a Notodontian 

 caterpillar found in some numbers on the l)lack l)irch in the White ]Moun- 

 tain valleys, which eats the leaf in precisely the same way as B. arthemis 

 does, and simulates its habits, even to taking its siesta on the denuded mid- 

 ril) of the apex of the leaf; but it makes no such ])ad of i-iff-raff next th.e 

 ragged part of the leaf as Basilarchia does. A similar relation is noticed 

 by Dorfmeister between the cater[)illars of Xajas populi and Xotodonta 

 ziczac of Em-ope, which feed in a veiy similar way on Populus. 



Life history. Edwards and myself have held different views regarding 

 the life history of this insect : and on reviewing both my own sources of 

 information, including numerous observations by many others as well as 

 by myself for many years, and the statistics he has so laboriously 

 gathered and published, I am inclined now to think that we were both of 

 us in part correct and in part wrong. As will have been seen above, the 

 caterpillar hibernates in mid-life, and awakes with the dawn of spring, in 

 the White Mountains in the latter half of May. It feeds for a few wrecks, 

 the chrysalis hangs for from nine to fourteen davs (nine to ten, Ed- 

 wards, ten to eleven, Scudder, ten to twelve, Saunders, "aiiout a fort- 



