80 SAMUEL H. SCUDDER. 



ward the ventral line (for the chrysalis lies upon its back), and thus 

 the ventral outline becomes straight, while the dorsal is strongly arched. 

 This condition of things is perpetuated and often intensified in the 

 next higher family, the gossamer-winged butterflies, which diflfer in 

 this respect from the typical butterflies only in the closer binding of 

 the girt around the middle. In the highest family, the brush-footed 

 butterflies, the girt around the middle is lost and the chrysalis hangs 

 suspended by the tail alone. We see therefore a regular progression 

 from the lower to the higher butterflies, in the loss first of the cocoon, 

 next of the girt ; and as if this were not enough, some of the highest 

 butterflies* have even lost the last remnant of silk and fallen to the 

 earth, where, amid stubble or in crevices in the ground, they undergo 

 their transformations without more ado. As if moreover to show that 

 this suspension of the chrysalis by the tail alone is a stage beyond that 

 of hanging by tail and girth, we have a clear proof that all the sus- 

 pensi, as Boisduval happily calls them, have passed through the stage 

 of the succincti, since the straight ventral surface of the abdomen, as- 

 sumed perforce by the succincti, when they left the cocoon stage and 

 became attached to hard surfaces, still remains in the chrysalis of the 

 hrush-footed butterflies, lohere it no longer serves any "puvpose, — as clear 

 and striking an indication that the suspensi outrank the succincti, as 

 that the pupa is higher than the larva. 



* Oenei's semidea, Agapetes Galathea, Nyiha Oircc and Eum&nis Semele. 



