HABITS OF THE LARVA OF BELLURA MELANOPYGA. 99 



beginning of the season it was no uncommon thing to find twenty 

 or more very young larvae, 3-4 mm. in length, on a single leaf. 

 The presence of so many very young larvae on one leaf and the 

 absence of larvae from a contiguous leaf is taken as evidence that 

 the eggs were laid somewhere about the infested leaf. 



During the mining period the larva works on the upper side of 

 the leaf. It usually cuts a somewhat circular hole, slightly larger 

 than itself, through the upper epidermis and penetrates into the 

 parenchyma, thus becoming a miner. There is no regularity in 

 the shape of the mine. Sometimes it appears as a winding tunnel 

 with a diameter about twice that of the larva; sometimes it is 

 digitate in appearance; and sometimes it resembles a "blotch 

 mine." The area included in each may vary to considerable 

 extent. Holes through the epidermis other than the original 

 entrance may occur anywhere throughout the length of the mine. 

 Mines are easily detected on the surface of a leaf since they soon 

 become whitish in appearance, due to the removal of the chloro- 

 phyll-bearing tissue. They may occur anywhere on the leaf. 

 Leaves were sometimes found in which the mines appeared to 

 extend either towards the midrib or towards the junction of the 

 midrib with the petiole but an examination of a large number of 

 infested leaves leads the writer to believe that no importance 

 can be attached to these cases. 



The initial entrance to a mine is usually surrounded by excre- 

 ment and a small quantity of finely masticated leaf tissue. 

 Microscopical examination shows that the latter is composed of 

 the broken fragments of the epidermal cells and numbers of the 

 peculiar idioblasts which are so common in the tissues of the yellow 

 waterlily. Examination also shows that the fragments of the 

 epidermal cells have not undergone digestion and furnishes 

 evidence for believing that the larvae never use the epidermal tissue 

 as food but simply remove it with the mandibles. The excrement 

 is greenish when first voided but gradually becomes lighter in 

 color until finally all color is completely lost. It is voided in the 

 form of small, somewhat oblong, uniform masses which tend to 

 remain together, forming chains of varying lengths. Micro- 

 scopical examination shows that it is composed of two kinds of 

 elements: (i) idioblasts, which form as much as 60 per cent, of 



