40 MASS. EXPERIMENT STATION BULLETIN 157. 



of food in their course has been previouslj^ exhausted by a l^rother miner, 

 or to pupate. They make their way to the pahsade parenchyma either by 

 mining over the edge of the leaf or by boring directly through the central 

 portion of the mesophyl. Rarely the newly hatched larva will bore its 

 way through in this fashion almost immediately, leaving no trace of its 

 existence whatsoever on the lower surface. 



The writer's curiosity was early aroused by this habit of the larva. 

 Of a number of reasons which at first suggested themselves to account 

 for it, only two were finally retained as being the more likely, viz., hght 

 and food. The question to be answered, then, was: Is the larva attracted 

 to the palisade parenchjina because of the more and greater degree of 

 light which that surface received, or because of the better food conditions 

 which the pahsade parenchyma affords? As regards the latter part of 

 the question, it is well known that the palisade cells composing the palisade 

 parenchyma are compact or close together, while the cells comprising 

 the spongy parenchyma are separated by comparatively large interspaces. 



To determine whether hght or food was the influencing factor, a number 

 of simple experiments were undertaken. In one series, the upper surface 

 of leaves which were infested by newly hatched larvae, which larvae had 

 not as yet made their way to the palisade parenchyma, were darkened by 

 being painted over with India ink. 



To guard against inconclusiveness of the fu-st series, owing to the pos- 

 sibiUty of the ink penetrating and proving repellent, another series of 

 similarly infested leaves were covered with black paper impervious to 

 light. In other cases similarly infested leaves were so fixed as to cause 

 them to remain in an upright position so that both surfaces received an 

 equal amount of hght. In still other instances such leaves were so fixed 

 as to reverse their surfaces, the true lower surfaces being turned up towards 

 the better light, the true upper surfaces being turned so as to receive less 

 light. In a few other cases areas directly in the course of larvae which 

 were already mining in the palisade parenchyma were darkened for the 

 purpose of determining the behavior of the larvae when the darkened areas 

 were reached. 



As a result of the above experiments it may be said that the influencing 

 factor is food supply. Light, however, did appear to be somewhat of an 

 influencing factor in some instances in the case of j^oung larvae. 



The mining (Fig. 10) as it appears on the surface of the leaves shows no 

 particular design. It appears as straight or irregular lines running trans- 

 versely or diagonally, but usually in a longitudinal direction, and often in 

 loops. This condition is still further comphcated in leaves which are in- 

 fested by more than one larva. In such cases the mining may be seen in 

 patches, or, as it very often happens, the entire leaf is mined. A favorite 

 course for mining is along the margin of the leaf. Within the mine the 

 course of the larva may be traced by a chain of black pellets of excreta 

 which it leaves in its wake. 



The mines, or better, perhaps, the channels, are as a rule within the 



