151 



transversely that it looks like a little knot. On opening it, 

 however, it is found to be essentially like that of one of the flat 

 group. As the larva leaves the mine to pupate, the mine never 

 becomes tentiform, but remains flat. It mines indifferently either 

 surface of the leaves of locust (Robinia). In the leaves of an 

 allied genus, Amphicarpaea, mines and larvae are found, which 

 are indistinguishable from those of L. ornatella ; the larvae have 

 the same life-history, and the cocoons and pup^e are indistinguish- 

 able. Judged by the mines, larvae and pupse, it is the same 

 species in the leaves of both plants. But the moth from Am- 

 phicarpaea belongs to a different genus. It is Leucanthiza 

 amphicarpeaeella Clem. The ornamentation differs from that of 

 all the known species of Lithocolletis, though perhaps it is nearer 

 to that of L. ornatella than to that of any other in some respects. 

 The antennae, which in Lithocolletis are carried back under the 

 wings in repose, are carried extended at the side in L. amphi- 

 carpeaeella ; the neuration of both pair of wings differs decidedly 

 from that of Lithocolletis, and is identical with that of Pliyllo- 

 cnistis vitigenella Clem., except that in the hind wings of X. am- 

 phicarpeaeella the median vein is furcate on the hind margin, 

 instead of being simple as in Phyllocnistis. In its other charac- 

 ters, however, it is near Lithocolletis. 



The mines of larvse of the cylindrical group are at first linear, 

 ending in a blotch, and they retain this character until after the 

 fifth moult of the larva. Up to that moult the trophi are those 

 of the flat group (fig. 2), and the larvae, though not flattened 

 like those of the flat group, are depressed and submoniliform, 

 while the feet, though better developed than in the flat group, 

 are small and feeble. The mine is of course only a separation 

 of the cuticle from the parenchyma, and as it is small, and on 

 the under side of the leaf (except in the case of L. rohiniella 

 and L. tiliaeella^ as before stated), it is likely to escape notice. 

 By far the greater number of mines and larvse of this group, 

 which one sees, have passed the fifth moult, when the larva has 

 become more cylindrical, though still a little depressed, and the 

 mines have been made tentiform. Hence the mines of this 

 group are usually described as tentiform, and the larvae as cylin- 

 drical. Before becoming tentiform, that is, before the fifth 



