108 USEFUL BIRDS. 



and probably drag the young larvae out of the smaller 

 twigs. 



It is extremely difficult and expensive to raise forest trees 

 in regions where there are no arboreal birds. The larvae 

 of several large nocturnal moths are among the most 

 destructive insects known. Mons. Trouvelot's statement 

 (pp. 30, 31) regarding the quantity of food eaten by the 

 larva of polyphemus illustrates the power for harm that 

 these creatures possess. The larvae of 8ainia cecropia and 

 Tropala luna are so gigantic and their rate of increase is 

 so great that they constitute one of the gravest dangers that 

 constantly menace our woodlands, yet we never hear of any 

 serious injury done by them in Massachusetts. Indeed, such 

 species, although large and conspicuous, are not often seen 

 except by entomologists and collectors of insects, who know 

 their haunts and habits. The main reason for their compar- 

 ative scarcity is indicated by Trouvelot's experience in rear- 

 ing the larvEe of Teleapolyphemus, the "American silkworm." 

 He had a tract of about five acres enclosed, and covered with 

 netting for the protection of these caterpillars. The vegeta^ 

 tion on this land as I saw it years afterward was largely scrub 

 oak and whortleberry or huckleberry bushes. Trouvelot says 

 that when he began rearing silkworms the bushes were of 

 about five years' growth, and it seemed as if there would be very 

 little to do after the place was enclosed and the insects put 

 in ; but he found that most of his time was occupied in de- 

 fending his treasures against the birds. The smaller of these 

 would push through the meshes or get under the edges, while 

 the larger ones would find some hole by which they could 

 enter. He says that he was obliged to chase them "all the 

 day long," as when he was pursuing them on one side they 

 would fly to the other and quietly feed until he reappeared. ^ 

 "Birds," he says, 'are the greatest foes of silkworms, espe- 

 cially the Thrushes, Catbirds, and Oi'ioles." He believes it 

 probable that in a state of nature ninety-five per cent, of 

 the silkworms become the prey of these feathered insect 

 hunters.'' 



Samuels tells us that Trouvelot was obliged, in self-defence, 



' American Naturalist, Vol. 1, p. 145. ^ Ibid., p. 89. 



