h52 



moult of the larvae, their mii'fes are always greenish, which aids 

 in preventing their discovery, and distinguishes them from mines 

 of the flat and ornatella groups, which are always brown, yellow, 

 or whitish, and therefore more conspicuous. 



The larvae, even in their ^arliest stages, differ from those of 

 the flat and ornatella groups more than those groups do from 

 each other. (Fig. 6, larva of L. rohiniella in its first stage.) 

 Still, even after its fifth stage, and until the latter part of its 

 eighth stage, a larva of this group is more properly described as 

 submoniliform than as cyHndrical. In its younger stages it is 

 more elongate, and is vertically thicker than a larva of the flat 

 or ornatella groups. Fig. 8c? gives the outline of X. rohiniella in 

 transverse section in its fourth stage. Its vertical thickness is 

 nearly equal to half of its width. The maculae are usually ob- 

 solete in these larvae, but they sometimes may be seen. Thus, 

 L. rohiniella mines indifferently either surface of locust leaves. 

 I have never found, out of the hundreds that I have examined, 

 a specimen from the under side of the leaf that had distinct 

 maculse, and but a small proportion of those of the upper side 

 show them. But sometimes they are found, at about the third 

 and fourth larval stages, as distinct as they ever are in the flat 

 larvEe, showing distinctly through the epidermis. At first I 

 thought that these maculate larva must belong to a distinct spe- 

 cies, but I have repeatedly bred them by themselves, without 

 having been able to detect any difference between them and 

 moths bred from larvae without maculae. At the fifth moult the 

 same changes take place in these larvae that take place in the 

 flat and ornatella group only at the seventh. The larvjB be- 

 come more cylindrical, the legs are better developed, the tropin 

 are as in fig 4, and the head is deflexed. A different sort of 

 mine is therefore needed. 



The mines are made in different ways by different species. 

 L. crataegella, lying on its back, spins its web across the inner 

 surface of the separated upper cuticle, whereby it is drawn, not 

 as in the flat group, into a single longitudinal fold, but into a 

 multitude of wrinkles like those in the mine of Ornix p7'7mivo- 

 rella, and the mine is deeper than that of a flat larva. L. rohi- 

 niella spins a somewhat dense web across the floor of its mine. 



