1224 THE BUTTERFLIES OF NEW ENGLAND. 



few, wliich devour iiinbelliferous plants, iill the known larvae live on trees 

 or shrubs, though of ver^' different groups. 



With the exception of some of the Parnassidi, which resemble the 

 Pierids, the chrysalids are always more or less angular and generally 

 rugose, the ocellar region advanced to form a bifid front, as in the Nym 

 phalidae ; like the Pierinae, they are suspended by the tail and a loose girt 

 about the middle, and some of the group of Parnassidi are also enclosed in 

 a cocoon of leaves loosely fastened together. Harris gives (Ins. Inj. veg., 

 264) the following description of the pupation of our native species : — 



It first spins a little web or tuft of silk against tlie surface whereon it is resting, 

 and entangles the hooks of its hindmost feet in it, so as to fix them securely to the 

 spot; it then proceeds to make a loop or girth of many silken threads, bent into the 

 form of the letter U, the ends of which are fastened to the surface on which it rests, 

 on each side of the middle of its body; and under this, when finished, it passes its 

 head and gradually woiks the loop over its back, so as to support the body and pre- 

 vent it from falling downwards. 



The division of the subfamily. Although only one of the two 

 principal subdivisions of this sultfamily occurs in our fauna and is treated 

 of in this work, it may be worth while to point out one or two of the 

 distinctions between them, as I have here and there intimated that such 

 tribal distinctions exist, though they have only been rarely and partially 

 recosfnizcd liv svsteniatists. The two trilics to which I refer are the Par- 

 nassidi and the Papilionidi, the latter only of which is found in eastern 

 America. To the former belong all those genera of Doubleday and West- 

 wood's work which follow their magazine genus Papilio. They differ 

 from the Papilionidi more particularly in the structui-e of the earlier 

 stages, some of the genera being related to the Papilionidi in certain 

 particulars much more closely thfin are others. In general, they may be 

 said to stand between the Pierinae on the one side and the Papilionidi on 

 the other. The ordinarily rounded form of their wings shows their 

 alliance to the Pierinae, while the uniform ncuration of the wings them- 

 selves shows that they cannot Jje separated widely from the Papilionidi. 

 With the latter they share the foliaceous appendage of the fore til)iae, the 

 four-bran(!lied median nervure of the fore wing, the excavated inner mar- 

 gin of the hind wing, the comparatively robust body, the more or less 

 arcuate antennae, and the absence of j)aronychia and pulvillus. 



So far as I have been able to see their eai-lier stages, which, excepting 

 for the mature caterpillar and chrysalis, is certainly to a very limited 

 extent, these tell the same story. The egg of Parnassius, the only genus 

 I have seen, and which, as well as the earliest stages of the caterpillar, 

 I know only through specimens recently sent me by Mr. Edwards,* 

 is in no pi'oper sense tiarate, but has a form more closely resembling that 

 of the Papilionidi, for they are broadly truncate and not greatly contracted 



•Statements made early in thi.s work iiroin part eiToueous, from distorted material. 



