1880.] 20 



and beautifully iridescent ; on the fifth day the embryo was distinctly 

 visible, on the eighth day the ocelli could be seen, on the tenth the 

 mandibles plainly, the segmental divisions and dorsal canal slightly, 

 and on the twelfth day the black head and plate on the second segment 

 showed most plainly, indeed, the slow undulating movement in the 

 alimentary canal could be distinctly traced ; about 10 o'clock the same 

 night, June 27th, they began to hatch, and the little larvae were all out 

 of the shells before next morning. 



" The colour of the larva when first hatched is creamy-white with 

 black head and plate. Fortunately I had growing plants of Solidago 

 virgaurea, Origanum vulgar -e, and other species, from which I at once 

 gathered leaves to see what they would take as food ; but at first they 

 were too restless, crawling upward and trying to escape from under 

 the glass cover of the pot ; but next day I noticed one larva had 

 attacked a leaf of Solidago by eating a small hole partly through it ; 

 subsequently other leaves were similarly eaten into, and by the begin- 

 ning of July they evidently preferred the golden rod and marjoram." 



With this account Mr. Jeffrey kindly forwarded to me a dozen of 

 the larvae on the 3rd of July, just as they had completed their first 

 moult, and were then not quite a tenth of an inch long, having a glossy 

 black head and collar-plate, a whitish pellucid skin, bearing blackish- 

 brown dots, and a broad yellowish-green internal vessel showing 

 through. 



On the 8th of July I found they had moulted for the second time 

 and become less transparent, excepting the belly which remained much 

 as before though slightly yellower, while the body above was darker 

 and rather dingy green, the head, the plate and dots shining brownish- 

 black. 



They lived contentedly but well apart from each other, and when 

 preparing for the third moult spun little hammock-shaped silk webs on 

 the under-side of the leaves, or between leaves slightly uniting the 

 surfaces. 



After the third moult, on July 17th, I was at first astonished at 

 not seeing the larvae, until I became aware that they were inhabiting 

 little cases, which they had made for themselves with portions cut from 

 the leaves of their food plants ; some of the cases were lying loose 

 among the leaves, though one or two at first were still adhering to the 

 leaves by a small part not cut away ; on turning out a larva two days 

 Jater I saw its back and sides were deeply tinged with dark purplish- 

 brown. 



Henceforward the larvae were not easily observed, as they were 



