186 THE entomologist's RECORD. 



means of the thoracic legs to gather up the thread, and so regain the 

 position from which it fell. As the larva is so small and the action so 

 rapid, I cannot yet say exactly how the thread is caught up, though I have 

 -attempted to settle this point by aid of a lens on very many occasions. In 

 order to see why the larvae let themselves fall, I carefully watched various 

 individuals as they crawled down the tree, and I noticed that when 

 they came to a branch, having a horizontal direction of growth, the 

 larvfB generally crawled along the lower side of the branch, and on one 

 occasion I actually saw a larva fall from the smooth bark. It fell only 

 to a depth of about half its own length and very soon pulled itself up 

 :again and recommenced its journey. This I have come to look on as 

 absolutely a mere accident, but it was curious that I should first 

 observe a larva fall in this way. The next thing I noticed was that 

 there were a number of ants running over the trunk and branches of 

 the tree. I soon found two ants descending the trunk, each carrying 

 a, larva of C. labitrnella in its jaws. Later I saw a third ant seize a 

 larva as it was crawling down the trunk, and the larva wriggled much 

 while being carried away. I tlien turned to a branch of the tree and 

 and watched both ants and larvfe. I several times saw the ants run 

 very close to the larvfe and the latter took no notice of them nor did 

 the ants seem to be aware of the larvas until they were almost in 

 <;ontact with them. At last I saw an ant actually come in contact 

 with a larva and the larva instantly dropped off the branch and fell to 

 A depth of about a foot. There it hung for a short time and then 

 began winding itself up again. I have not the slightest doubt that 

 the larv89 I had previously noticed falling from the tree to about a 

 foot's depth had also been attacked or irritated by ants. On the other 

 hand there is no doubt in my mind that scores of these larvfe do reach 

 the earth hanging on their threads. I have seen them do so. The 

 process of winding themselves up again when they have once fallen a 

 foot or more occupies a very long time and must be tedious to the 

 larva. They frequently make a second fall while ascending, but 

 whether this is owing to the fatigue, or to an accident in the winding, 

 I cannot say. I fancy that after a second fall they rarely, if ever, 

 regain the tree, and therefore sooner or later fall to the earth. When 

 there is a breeze blowing the larvfe sway about on their threads and of 

 course often get entangled with the threads of other larvfe. That the 

 larvae reach the earth by descending on threads of their own silk is 

 perfectly true, but I maintain that this is not a premeditated action, 

 that it is not done without some kind of compulsion, that it is 

 always accidental, that is to say that the larva does not descend 

 from its own subjective idea, but on account of some objective 

 interference with its natural procedure. I captured some of 

 the ants I have mentioned and sent them to Mr. A. H. Hamm, who 

 liindly identified them as Lasiim niijer, L., a common species in most 

 gardens. Since 1908 I have noticed larv;^ of various species, especially 

 those of the Geouietrina, hanging by threads, and they were always 

 likewise trying to ascend. But of course many of these larvae were 

 not full fed and were trying to get back to the food. Such, however, is 

 never, I believe, the case with C. laburnella, as they do not leave the 

 mine till they are full fed. Many leaf-mining larvje have the power of 

 moving from one leaf and making a fresh mine in another leaf, but 



