68 LEAF-MINING INSECTS 



leaves this mine when about 2 mm. long, and, entering 

 between the cuticles of the leaf of the common morning 

 glory at another point, makes a blotch mine but it either 

 leaves the posterior segments of the body extruded or it 

 returns to the entrance for the purpose of ejecting frass. 

 Parectopa robiniella makes a digitate mine on the upper 

 surface of the leaf of black locust but it enters the leaf from 

 the lower side, making a very small mine there and then 

 passes up through the plane of fibrovascular bundles to 

 mine under the upper cuticle. It keeps returning to the 

 hole into the lower side mine to stuff its frass down there. 

 Sometimes its cast skins may be packed in with the remain- 

 der of its waste. Those superb miners, the Gracilaridae, 

 do not eject excrement from the mine but the ingenuity 

 some of them show in otherwise getting it out of the way is 

 interesting. For example, Lithocolletis lucetiella, mining 

 the lower surface of basswood trees, during its early instars 

 while it has sap-feeding mouthparts, carefully shears loose 

 the lower epidermis, feeding on the liquid contents of the 

 cut cells. The excreta of this period are so small in amount 

 as to be unimportant. After the third moult when tissue 

 feeding mouthparts are acquired the caterpillar does not 

 extend the mine but goes back over this rectangular vein- 

 bordered blotch, eating as it goes every vestige of paren- 

 chyma and leaving a splendid dissection of the fibro- 

 vascular network. Now the amount of undigested material 

 is increased very much and the problem is to dispose of it 

 with the least possible contamination of the food supply, for 

 by taking all the parenchyma in that scant area there is 

 none too much to bring the larva to full growth. This is the 

 solution: The larva begins eating at the very edge of the 

 mine and eats all the way around the edge first and then 

 gradually in toward the middle. By the time the first 

 frass pellet is ready, there is a narrow space at the edge 

 where the larva has just been eating that is entirely stripped 



