84 PROCEEDINGS OF THE AMERICAN ACADEMY 



The excellent differential characters separating this genus from 

 other Attacina? are now defined, and the same will apply to the larva? 

 of the third stage of Platysamia, Callosamia, and Actias, as well as 

 Samia (Philosamia). 



The following descriptions apply to two individuals specially observed 

 during this stage. 



One (A) in Stage II. was seen to cast its skin, July 5, at 11.15 to 

 11.30 a.m. The head was pale greenish yellow, like a peach, but 

 without the reddish pink tinge. (A. luna appears to permanently 

 retain the greenish tint.) The thoracic legs are greenish. All the 

 tubercles are lemon-yellow, the short bristles on the thoracic tubercles 

 black, those on the abdominal segments turning black. The long 

 whip-like hairs are white. The V-shaped band on the edge of the 

 suranal plate is a deep labradorite-azure. The lateral stripes are not 

 yet very distinct. The spiracles are deep orange. 



At 12.30 p. M. the head had turned almost chestnut-brown, and by 

 1 P. M. was of the normal dark chestnut-red hue. 



Before casting its skin, it spins a thin carpet of silk threads, to which 

 it clings with its crotchets while in the process of exuviation. 



Another caterpillar (B) in Stage II. about moulting was first noticed 

 at 11.30 A. M. The head was small, about half as large as in the next 

 stage, pressed forward ; the prothoracic segment above has a large 

 yellow patch extending back to the next segment. The region is about 

 half as wide as the whole segment, being that portion situated behind 

 the two middle dorsal tubercles, and the brown membrane or neck 

 connecting the head and the succeeding segment is tense. Now all 

 the tubercles are deep orange-red, while there are no fine white hairs 

 arising from the thoracic tubercles, and those arising from the abdom- 

 inal tubercles are much shorter, nearly one half, than in the next 

 moult (A). It fastened its crochets in the silk carpet it had spun 

 previous to the beginning of the process of exuviation, so that the 

 convulsive movements of the head and thoracic segments may not 

 cause it to fall over while in the act of throwing off the old skin. The 

 head is about one third, and almost one half, larger after moulting 

 than before. t 



Xow and then before the skin splits, and is cast off, the larva was 

 observed to make a series of convulsive movements of the head and 

 thoracic region. 



It finally cast its skin between two and three o'clock p. M., and this 

 individual looked like A when I first saw it. 



This larva also was observed resting with its head and thoracic 



