139 



pressed; the jaws project in the same way beyond the head, 

 and can neither be elevated above nor depressed below the axis 

 of the body; this is equally true of the larvae of the other 

 groups in their earlier stages. Such a larva placed upon the 

 surface of a leaf could never enter it, but must perish ; but, 

 when the egg is so imbedded in the surface of the leaf, and so 

 firmly attached to it, the larva, in passing obliquely out of it, 

 necessarily enters the leaf. This affords a sufficient reason why 

 a larva once removed from its mine always dies, so long as the 

 structure of the head and moutli-parts retains this character 

 (figs. 1 and 2); though no good reason, apart from the crea- 

 ture's instinct, can be given why it might not, after the change 

 occurs, which takes place in all at some period, and after the 

 trophi become like those of ordinary larvae of Lepidoptera, as 

 shown in fig. 4, re-enter the leaf, and form a new mine, as do 

 the larvae of many other genera (e. ^., some species of Gracil- 

 aria, Ornix and Laverna), or feed externally, like the greater 

 number of lepidopterous larvae. But, as a matter of fact, it 

 never does so. Dr. Clemens has stated that the larva of L. 

 crataegella, when feeding in leaves of the Avild cherry (^Prunus 

 serotina'), sometimes leaves one mine and makes another. If 

 this is true, its habit is unique in the genus ; but, though I am 

 very familiar with this species, I have not been able to confirm 

 Dr. Clemens' statement, and I have found reason to think that 

 he was in error. Ornix prunivorella Cham, was unknown to 

 Dr. Clemens. It mines the leaves of Prunus serotitia, and 

 its mine cannot be distinguished from that of L. crataegella 

 Clem., and, like other species of Ornix, it does leave one mine 

 and make another. I think it at least probable that Dr. Clem 

 ens mistook the Ornix mine for that of L. crataegella. 



To return from this digression. After leaving the egg, the 

 mines of all the species of Lithocolletis that are known to me, 

 as well as those of many other genera of Tineina (^e,g., Phyl- 

 locnistis), are at first linear and confined to the surface, upper or 

 lower, as may be the habit of the species. Larvge with the 

 trophi as in fig. 2 (e.g.^ Phyllocnistis, Lithocolletis, and very 

 young Gracilaria) simply separate the epidermis from the pa- 

 renchyma, and do not eat the latter. The linear part of the 



