Ill'; lANAIHAN KNTMMdI.dClsT, 



the liasL- of the leaf, and s])ins tlireads across it. Tt is not difficult to 

 bring the edges together, as the K'al is hut partially opened, the edges 

 being somewhat curled in. Within a few hours, the whole leaf will be 

 found to be stitched in the mantn-r which 1 will describe at one larval 

 stage huer. Hut an opening is not left at the tip for the Iar\a to thrust 

 out its head : nor does it begin at the lip and eat downward. That would 

 a])|)lv to the larva from second moult onward. I'he \ouiig one eats the 

 substance of the leaf within its case, at some distance from the ti]). leaving 

 the framework untouched. It makes its resting place cpiite close to the 

 base, and there it will remain till after it has ])assed the first moult. The 

 case at this stage is never so eaten that it does not afford ])rotection to the 

 larva, and i)rotection at the youngest stage is most important, as it is then, 

 if ever, that the little i< hneimion-fl\- deposits its egg. The net-work of the 

 frame of the leaf would seem sufficient to keeji out man\- sorts of 

 enemies. Most of the feeding is done at night. .\t last, some morning, 

 the tenant will be found to have deserted its case, and to have shut itself 

 up in one of the second pair of leaves, and it is now past the first moult. 

 I experimented on several larva.' just past this moult. ( )ne was ])laced on 

 the terminal leaf of a plant set in a flower-])Ot in mv room. A few minutes 

 later, it had descended to the secoud pair, and taken possession of one of 

 them. Its hrst movement was to gnaw nearly through the mid-rib (|uite at 

 the base of the loaf and also to cm a hole on either side, the rib at the 

 break. The leaf was thus made to droop several degrees, and at the same 

 time the sides moved closer together. Then s])inning began. Threads 

 were laid oblicjuely across die o])en s])a( e. one end fiistened to the base of 

 the leaf on one side, the other a little above the base. After the edges 

 had l)een bound for a short distance in this way. a change was made and 

 threads were spun directl) across, and over the first ones, begin- 

 ning at the base. One set of threads drew the edges, the other held them. 

 Then the oblique threads were laid again, and the cross threads, and so 

 on, alternating, till at 90 minutes from the start, sometimes working, some- 

 times resting, the little creature — .2 inch only — had closed half the 

 length of the leaf It had also spun some coss threads within, which 

 would hel]j hold the sides in jjlace. Next morning the leaf was com- 

 pletely closed, and to get a view of the larva I had to slit the case with 

 scissors. This was the usual mode at the younger larval stages, only that 

 at the first, on the terminal leaf, the rib was not bitten nor were holes 

 made at the base. 



