THE TROILUS BUTTERFLY. 267 



hatched it is slate-colored above, with a black spot like an 

 eye on each side of the third segment, below and behind 

 which is a large and long white spot, and the top of the 

 eleventh segment is white. After changing its skin, it be- 

 comes of a pale brownish olive color, the white spots dis- 

 appear, and on the top of the back we find two rows of 

 minute blue dots. When fourteen or fifteen days old it 

 changes its skin and its colors again, the back becoming pea- 

 green, with blue dots, the sides yellowish, and the head, 

 belly, and legs pink ; there is a transverse black line on 

 the top of the first segment, and there are two large orange- 

 colored spots on the fourth segment, and two of the same 

 color, with a black centre, on the third segment. The cat- 

 erpillar retains these colors from ten to sixteen days, increas- 

 ing greatly in size during this period, and finally attains to 

 the length of two inches or more. It comes to its full 

 growth when about four weeks old, and then eats no longer, 

 but, deserting its leafy habitation, it seeks a suitable place in 

 which to undergo its transformation, previously to which it 

 casts off its green coat, and appears in one of an ochre-yellow 

 color. It then suspends itself in the same way as the cat- 

 erpillar of the Asterias butterfly, and within two or three 

 days after its last change of skin it moults again, and be- 

 comes a chrysalis. 



The chrysalis is generally of a pale wood-color, smoother 

 than that of the preceding species, and with rather longer and 

 sharper ear-like projections. The chrysalids, which are pro- 

 duced from caterpillars hatched in August and September, 

 remain unchanged through the winter, and are not trans- 

 formed to butterflies till the middle of the following June. 

 It is possible that these butterflies may lay their eggs so early 

 as to produce a brood of caterpillars in the summer, and these 

 may come to their growth, and pass through their transfor- 

 mations, before September ; but I have only found the cater- 

 pillars towards the end of summer. I once discovered them 

 on the leaves of the lilac, on which they appeared to thrive 

 ^uite as well as on the sassafras. 



