EREBIA III. 



fore wings, as in Fig. 6. The outcome was a single male, true type Uj/qjsodea, 

 scarcely banded beneath, out of chrysalis 12th May, 1888. This is shown on 

 the Plate, Fig. 1. The eggs were laid 8th July and hatched 2Uth. On 2oth, 

 there were seven young larvoe. The first one passed first moult 27th July, the 

 second moult 4th August, the third moult 28tli August. The others lingered 

 in their stages, but by 15th September, five had passed third moult. They were 

 at all times kept out of doors, and cool weather now coming on they ceased 

 feeding gradually. On 29th October, I found but three larva', all in lethargy, 

 two of the five having disappeared. In November, 1 took them to Clifton 

 Springs, as before stated. On 5th April, 1888, I received them again, two alive 

 and wide awake as I opened the box. These w'ere at once placed on grass, and 

 in five minutes thereafter were feeding. On 15th April, one pas.sed its fourth 

 moult, the other on 17th. On 30th April, 1 noticed that one was bringing 

 together leaves of the grass and forming a sort of loose C3'linder. It was in the 

 middle of a pretty dense tuft, the leaves of which were three to four inches long. 

 On one of these the larva rested, holding by its prologs, and was spinning a few 

 threads and drawiui^ the leaves down and about it. Before night the inclosure 

 was complete. Nearly a score of leaves were held, mainly by threads at top 

 and bottom, that is, above and below the larva, making a pretty close covering, 

 but open by spaces so that most of the side, and the head, were exposed to view. 

 The larva rested head up, back arched. Mr. Scudder, in But. N. E., has well 

 characterized this structure as " an imperfect cocoon." Pupation took place on 

 the night of lst-2d May. Some hours after, when the chrysalis had hardened, 

 I cut away the leaves one by one. There were fifteen of them, and the pupa 

 rested upright, its lower end one and a half inch above ground, in an angle 

 formed \>y three leaves. As I cut one of these, it turned over and fell, showing 

 itself to be unattached. In fact, the cremaster was found to be furnished with 

 but sti-aight bristles, very short (Fig. g'"), and there were no hooks by which 

 attachment could be had. As before stated, a male of the typical form came 

 from this chrysalis on 12th May. 



I had received young larvaj of this species from Mr. Bean, at Laggan, 25th 

 July, ISSG, they having hatched e7i route. I lost all Init one of these. It passed 

 first moult 2d August, the second moult 10th August, and soon after became 

 lethargic. I kept it at Coalburgh, and brought it into the house middle of Janu- 

 ary, 1887, apparently healthy. But a month later it unaccountably disappeared. 



Mr. Bruce had also sent eggs from Colorado which reached me 28th July, 

 1886. From these, five larvae reached second moult, and went into hibernation, 

 and died during the winter. Fi-om the behavior of the different lots of larvsB, it 

 appears that hibernation may take place at either second or third moult. 



