March 1807.] Davis : Intelligence shown by Caterpillars. 43 



tended beyond the petioles of the leaves. During the storms of winter 

 most of these unsecurely supported cocoons are also broken loose and 

 complete their descent to the ground. We have found in early August 

 an American silk-worm cocoon attached to the side of a house, about 

 five inches from the ground, and on the loth of last January, Mr. Chas. 

 W. Leng and I, while walking on the ice in a Staten Island swamp, 

 discovered one firmly woven to a forked branch of a rose bush, that 

 stood in the water. It was as well secured to the forked branch as a 

 Cecropia cocoon would have been. Of course to have fallen off with 

 the leaves would have resulted in the death of this particular Folyphei?ii(S, 

 and we presume that it was in some way the realization of this fact that 

 caused the caterpillar to attach the cocoon so securely. 



On the bushes and small trees that grow in the water on the margin 

 of Silver Lake, on Staten Island, we have found a Luna moth cocoon 

 and also an Angulifera cocoon firmly attached to branches. It is the 

 habit of both of these species to construct their cocoons on the ground, 

 but not being able to do so in the cases cited they did the next very 

 best thing possible. 



The above are perhaps not quite as interesting cases of the care 

 taken by caterpillars to preserve their cocoons and themselves as the 

 one mentioned in the " Proceedings of the Natural Science Associa- 

 tion" some years ago under the caption of " Woodpeckers and Cecropia 

 Cocoons." As is well known, Cecropia caterpillars spin their cocoons 

 in a variety of places, often on fences, sometimes at the base of elder 

 bushes and sometimes at the ends of swaying branches, when the food- 

 plant happens to be a tree. The cocoons spun near the ground are 

 often devoured by mice that gnaw through the silken coats to the edible 

 pupa within. Those placed on tree branches are more safe from the 

 attacks of mice, but are liable to be eaten by woodpeckers. On the 

 14th of January, 1888, 1 saw a Downy Woodpecker investigating a Cecro- 

 pia cocoon in a white maple, the woodpecker thrusting its bill in and 

 pulling it out of the cocoon quite frequently. After a while it flew to 

 another cocoon a few feet away, but it being on such a small branch it 

 was unable to successfully pick it open as the branch swayed up and 

 down. It was then plain what a great protection it was to the insects to 

 place their cocoons near the branch ends, though no doubt they are 

 sometimes killed by the swaying of these branches during a storm. 



When the woodpecker was gone, I cut the cocoon off, and found a 

 small hole in its side quite near the branch, where it was easiest to drill 

 because the silken fabric gave way the least to the strokes of the bird. 



