148 



Neither can I conceive any good reason why moths disclosed 

 late in the fall might not winter, and indeed the moths of L. 

 rohiniella and L. salicifolidla do ; but I have never met with 

 any others hibernating, and indeed I have not seen L. salicifo- 

 liella later than November. There are, however, some facts 

 about the hibernation of tlie species which need further explana- 

 tion. Tims, I have known L. tuhiferella Clem, and L. aesculi- 

 sella Cham, to pass into their last larval state in the middle of 

 August, and to remain in that condition until late in the fall, 

 when my last observations were made upon them, and in such 

 cases there was abundant warmth and time for another brood 

 before the fall of the leaves. 



It now remains only to notice the points in which the three 

 larval groups differ from each other. The larvae of the flat 

 (second) group are the flattest lepidopterous larvfe known to 

 me. Fig. 7 represents the larva of X. coryliella in its first 



- CD 



Fig. 6. Fig. 7. Fig. 8. 



E.\;PL.\NATION OF THE FIGURES. 



Fig. 6. Larva of L. rohiniella, first stage, 0.63 mm. lo'ig. 



Fig. 7. Larva of L. corylitlla, Cham., first stage, 0.56 mm. long. 



Fig. 8. a. Outline of L. coryliella in transverse section at the fifth stage, b. Out- 

 line of L. ornatella in transverse section at the fifth stage, c. Outline of L. ornatella 

 in transverse section at the eighth stage immediately after the seventh moult, d. Out- 

 line of L. robiniella in its fourth stage. 



stage, 0.56 mm. long ; fig. 1, an older larva of the same group ; 

 fig. 3, the same towards the latter part of its last stage ; fig. 8a, 

 the outline of one of these larvae in transverse section at about 

 the fifth stage, when the vertical thickness is scarcely one-fifth 

 of its width. The " macidae " of Dr. Clemens are usually only 

 transverse rings, as shown on each segment in fig. 1 ; they 

 appear at the second stage and persist throughout the larval life, 



