9 [Jamtavy, 



again, and in instances too whei'e the larva could have been batched 

 little more than twenty-four hours. That the egg is laid at the point 

 where the larva enters I have little doubt, but whether the empty- 

 shell scales off, or furnishes the larva with its first meal, or is present 

 all the while and wants better eyes to see it, I cannot say. The first 

 explanation may be the right one, as there is always at the point of 

 entrance some tearing and contraction of the cuticle, and even of the 

 whole thickness of the leaf when the mine happens to start near the 

 edge. All of them enter from the under-side ; those that commence 

 to feed very early in the year begin close to the edge of the leaf, as 

 this is the part the parent insect can best reach when laying her egg 

 in the just opening bud, while salopiella and Sparmannella, that are 

 later in both the perfect and preparatory stages, enter more in the 

 body of the leaf. The last-named two also commence with a long 

 Nepticiila-\\\Q gallery, whereas, in the others that enter near the edge, 

 the gallery is so short and twisting as practically to be non-existent. 



The larvae grow rapidly, and change their skins three times, the 

 moulting period being known by the withdrawal of the creature from 

 the margin of its mine. They are tough skinned larvae, stiff and 

 clumsy looking, with stout thoracic and slender abdominal segments, 

 deeply incised ; and when in their mines, where the details of their 

 structure cannot be ascertained, they bear a striking resemblance to 

 certain footless Coleopterous larvae, the resemblance extending even 

 to the thread-like form of the frass. In the summer of 1888 I found 

 in elm leaves deserted mines full of thread-like frass, which, in my 

 then ignoi-ance, I unhesitatingly ascribed to a Micropteryx, but re- 

 peated search this year having failed to confirm the conclusion, I am 

 now disposed to think they were merely the work of one of those 

 beetles, that are often abundant on the elm in early spring. 



But to return to our Micropteryx. The head is very small, flat, 

 and pointed ; its posterior lobes are prolonged backwards to an ex- 

 cessive degree, making the notch at the back of the head very wide 

 and deep. This development of the lobes is present in all mining 

 larvae, and affords a powerful leverage to the muscles that move the 

 head from side to side, which is the movement these larvae employ in 

 feeding instead of the usual up and down one of external feeders. 

 Segment two is much wider than the head, and overlaps it ; it is shallow 

 in front, and deep behind where it joins segment three. Segments three 

 and four are about as wide as segment two, and very massive ; instead of 

 being cylindrical in section, each is roughly square, in consequence of a 



