224 [March, 



little below the tail, the bendiugs and contortions were not interrupted 

 by my interference, nor was the effort to reach the silk in the least 

 abated. Held firm by the stretched ligament, which was in plain 

 view, the body rose, and the tail, which had got well outside the padded 

 skin, and was before complete extrication bent backward, now bent 

 forward, and by the upward " swing was brought exactly to the silk. 

 Several times as I was Uf ling, the skin and chrysalis together were 

 dislodged, and fell into my hand. Then by drawing the skin back the 

 ligament was exposed and it was distinctly seen that it was attached 

 to the chrysalis by the pointed ends of the ridges before mentioned, 

 and that there was no other connection between skin and chrysalis. )) 



After the booklets of the tail are caught in the silk, the chrysalis 

 whirls one way and then the other, the last segments actively twisting 

 and screwing in order to fasten the booklets more securely. This 

 movement does not seem to be made for the purpose of rupturing the 

 membrane or for getting rid of the old skin especially, for I noticed 

 that whenever the skin parted and fell just as the silk was grasped, 

 as did sometimes happen, the same whirling, and all the movements 

 usually seen, followed. It is a wonderful exhibition, and the last act 

 is beyond my comprehension, — namely, the rising of the chrysalis with 

 no external aid save what comes from the ligament. I can only state 

 the fact. 



When the rupture of the skin of the caterpillar of interroqationis 

 first takes place, and the mesonotum is made to appear, this organ is 

 pressed down and flattened, but in a short time, and before the trans- 

 formation is completed, it swells out, and becomes nearly as large and 

 as prominent as it ever will be ; the head case is pushed forward on 

 the thorax and jammed in, so that on first issuing, the chrysalis is 

 truncated at the anterior side of the mesonotum. "When the skin ia 

 thrown off, the chrysalis hangs limp and distended, like a long cone, 

 with no prominences except the mesonotum. Presently the segments 

 shorten and become broader, the ends of the wing cases creep nearer 

 the tail, the tuberculated points on the abdomen swell out, the head 

 case pushes up, with its palpi cases, and in course of half an hour the 

 final and characteristic shape is assumed. The change in these respects 

 is nothing like so striking in Grapta as in Limenitis, where the chry- 

 salis is greatly hunched and displays a prodigious mesonotum. In this 

 case the chrysalis is at first as limp and shapeless as in Grapta, but 

 reaches its proper form in the same way ; the segments contracting 

 and the processes growing and maturing as one looks at them. 



The transformation of Arcliippus presented a close resemblance, 



