1072 THK HUTTKUFI.IES OK NKW F.N(;i-A\I) 



they had been placed in tlie cellar. The hiitterHies remain on tlie wing in 

 northern Florida until November and, by Dr. ('hapnian's notes, appear 

 affain earlv in Fcbruarv, but onlv in scantv nund)ers. oidv two or three 



Of ... 



spechnens being seen any day until May, about the middle of which month 

 fresh individuals appear ; by the end of this month they become abundant 

 and continue so throughout the rest of the season, there being a notable 

 accession of fresh material about the second week in August. These two 

 dates then, the middle of May and the middle of August, doubtless mark 

 the advent of fresh broods. As Edwards says that in West Virginia "the 

 butterfly is jjresent in several successive generations and in ()verlap))ing 

 broods from early in May to winter," and as, according to his figures, it 

 takes less than a month for the passage of the preparatory stages, there is 

 without doubt a brood between these two. Maynard says that there is 

 evidently a fresh brood in Florida in November and the same thing appears 

 to be true in Georgia, judging from specimens received from Dr. Oender ; 

 80 that it appears most proi)al)le that there are at the least four regular 

 broods. These become confused toward the end of the season ; thus Ed- 

 wards says : "On one day in September ... I cut a brancli of wild senna 

 (Cassia) on whicii at the moment were newly laid eggs . . . larvae in every 

 stage of growth and a butterfly . . . just emerged and still resting on the 

 empty shell of its chrysalis" (Can. ent., vii : 1W2) : and he afterwards 

 adds "there were forty larvae on one stem." 



The <lurat ion of the cffn; state in midsummer. accordin<r to Edwards, is 

 onlv two or three days, the different larval stages about the same, and the 

 chrysalis hangs eight days in July and August. Abbot gives the same 

 period for the chrysalis in Georgia. In Cuba, according to Gundlach, the 

 chrysalis hangs only six days in September. Riley tells me that in Mis- 

 souri he found it to last from five to eigiit days in September. 



The butterfly, according to Doubleday, fre(pients open plains near forests 

 and clover fields, in the latter of which it associates witli Eurymus philo- 

 dice and Zerene coesonia, wiiose flight it resembles more than it does that 

 of its nearer allies. Maynard, however, calls its flight sluggish, and 

 Gundlach states that it is very direct, and he adds that it assembles in 

 flocks on moist mud like otiier Rhodoceridi. 



Parasites. With the caterpillar of this l)uttcrfly. Riley found in Mis- 

 souri the cocoons of a parasitic hymenopteron. Apanteles cassianus, up to 

 the ])resent its only known sj)ecial enemy. 



Desiderata, ^^'e have endeavored to trace an outline of the history of 

 this biittei'fly but upon very Insuflicient data ; the sequence of the broods 

 needs following upon the spot, where also it should be determined whether 

 the insect hibernates otherwise than as a butterflv. and if so, what propor- 

 tion of tiie hibernators are butterflies. Do the October and November 

 butterflies ever lav ewe's the same season ? The earliest larval stajjc is in- 



