THE AMERICAN ENTOMOLOGIST. 



257 



developed articulated feet on the thoracic 

 segments, and its pro-legs are armed 

 with the circle of hooklets, as in ordi- 

 nary caterpillars. In the flat group of 

 larvge, however, the jaws are not only 

 greatly aborted and functionless in the 

 last two stages of larval life, but the tho- 

 racic feet, as well as the ventral ones, 

 are membraneous, and neither set is armed 



[Fig. 136.1 



First form of trophi and antennas in Lithocolletis guttifini- 

 tella (after Chambers). 



with either claws, or the circlet of tenta- 

 cles. These differences of structure would 

 seem to be correlated with the differences 

 of habit above mentioned; but if so, what 

 shall we say as to the larvK of the third 

 group of the same genus, comprising the 

 majority of the species, and in allusion to 

 the form of the larvce called the " cylindric- 

 al " group ? These larvae are flat in their 

 first three stages, as are those of ornatella 

 and the flat group, in their first five stages. 

 But at their third molt they undergo the 

 same changes of form and of trophi and 

 feet that ornatella does at its fifth molt. 



First form of trophi in Phyllocnistis vitifoliella 

 (after Chambers). 



and are, therefore, structurally as capable 

 of leaving the mine and crawling away as 

 is the larva of ornatella. Yet, they not only 

 do not do so, but continue to feed in the 

 mine, and pupate there. This, however, 

 is apart from my present purpose. The 

 fact, however, indicates that the develop- 

 ment of the jaws and feet is independent 



of the habit of leaving the mine, and is also 

 independent of their habits in the mine. 



I have already referred to the form of 

 trophi, shown in Fig. 126, as different from 

 the form shown in Figs. 124 and 125. Fig. 

 126 represents what I call the first form in 

 allusion to the time of larval life when it is 

 exhibited; Figs. 124 and 125 represent in 

 the same way the second form, and is, as be- 

 fore stated, the form found in ordinary cat- 

 erpillars when they leave the eggs; and, in- 

 deed, in all caterpillars, so far as is known, 

 except in the earlier stages of the larvae of 

 the genera Fhyllocnistis, Lithocolletis, Leu- 

 canthiza, Coriscium, Gracilaria and Ornix ; 

 and it is found in the later stages of them. 



[Fig. 128.] 



Phyllocnistis vitifoliella : a, head of last larval stage ; 

 b, head of immature pupa pressed of the larval head. 



The difference consists not only in the 

 different size, form and degree of develop- 

 ment of the different organs, but in the 

 presence or absence (apparently) of some 

 of the organs themselves. Thus, in the first 

 form (Fig. 126), only labrum, mandibles 

 and labium have been detected; the spin- 

 neret is very rudimentary, and there is no 

 trace of either maxillary or labial palpi, or 

 of maxillae.* In the second or ordinary 

 form not only are the labrum and labium 

 greatly altered in form and size and rela- 



* In the larvae of the flat group, in which the trophi of the 

 first form attain their full size at the end of the fifth stage, and 

 in which they are larger than in any other of the genera or 

 species mentioned, their width is only l-150th inch; length 

 about l-300th inch; and vertical thickness not more than 

 l-2000th of an inch. The front margin of the labrum is scal- 

 loped and fie.xed downwards in the middle (Fig. 126), so that 

 the teeth of the mandibles appear in a focal plane above the 

 labrum, which is dentate and ciliated in some species. These 

 facts, with the result of my first dissections (made with needles), 

 induced me to think, and to state in the address above referred 

 to, that the ma.xillae were united above the labium, forming 



