141 



ornatella, so tliat they arc unfit for crawling, when the larva is 

 removed from the mine, in all except the last stage of L. orria- 

 tella, which voluntarily leaves the mine and crawls away to 

 pupate. No good reason can he given why the cylindrical 

 larvae, after their fifth stage, when the feet are apparently large 

 and strong enough for use, seem unable to crawl when removed 

 from the mine. When out of the mine they apply the spinneret 

 to the surface on which they rest, and spin a thread fit to afford 

 a secure foothold, as do most crawling larvae, but they are luiable 

 to crawl, and yet if, while tumbling helplessly about, the true 

 feet happen to touch a part of a mine from which the upper 

 cuticle has been removed, the larva at once drags itself upon 

 the mined portion, and then crawls actively enough, without 

 attempting to spin a thread for a foothold. 



From each side of each segment project three hairs, just above 

 which are two other shorter ones. These hairs are found in all 

 the groups, and persist throughout the larval life. 



I have mentioned above the Phyllocnistis-like character of 

 the tropin and of the mine in the earliest stages of Lithocolletis. 

 Indeed, while this character of the tropin is retained, this char- 

 acter of the mine results as a necessary consequence. The 

 larva can only feed straight on or turn to one side or the other, 

 merely separating the cuticle from the parenchyma, and cannot 

 deflect the head so as to eat out the latter, as a mining larva 

 with the head deflexed and the trophi as in fig. 4 would do. 

 The form of trophi in fig. 2 is found in the earlier stages of 

 some other genera besides Phyllocnistis and Lithocolletis, and in 

 such cases the mine is usually linear and is always a mere sep- 

 aration of the cuticle from the parenchyma. It need not be 

 linear. Indeed, in Phyllocnistis ampelopsiella, though the mine 

 is, strictly speaking, linear, yet it winds about from the midrib 

 to the margin and back, between the veins of the leaf, until the 

 entire cuticle in the mined portion is separated, and the mine 

 becomes a blotch. In Lithocolletis and many other genera, the 

 mine always becomes blotch-like. Frequently, as in L. orna- 

 tella, the blotch obliterates the linear part of the mine, but in 

 other cases, as in L. celtisella, the linear part is long, and the 

 blotch is at one end of it, like the mines of some Nepticulas. 



