48 THE RELATION OF SPARROWS TO AGRICULTURE. 



placed in the cage met with a similar fate. Stinging insects were 

 not given the song sparrow, bnt some English sparrows to which tliey 

 were offered refused to touch them; which seems to indicate that 

 sparrows are able to distinguish between stinging and stingless insects 

 that resemble each other closely. Spiders were eaten by the song 

 sparrow with a relish that helped to explain the frequency with which 

 they are found during stomach examinations. A lace-wing fly, an 

 insect of the most nauseating odor, was also eagerly devoured. Stink- 

 bugs of the genera Brochymena and Euscliistus were refused; but 

 smaller bugs of the families Reduviid?B and Lyg?eid8e, though strong 

 scented, proved to be exceedingly palatable; and another ill-smelling 

 bug, the common water-strider (Hydrotreclius sp.), was eaten without 

 hesitation. Small brown species of leaf-hoppers and a green form 

 [Diedrocepliala) were offered to the song sparrow and were quickly 

 snapped up, showing that some device other than ill flavor probably 

 protects them ordinarily, since leaf -hoppers, though often exceedingly 

 abundant where sparrows occur, seldom, if ever, form any significant 

 part of the food. 



Several experiments were made with Lepidoptera. A salt-marsh- 

 caterpillar moth {Leucarctia acrcBci)^ a white and yelloAV insect warn- 

 ingiy colored and possessing a powerful odor, was given to the song 

 sparrow, but was refused. Then a yellow swallow-tail butterfly 

 (Papilio turnus), also a warningly colored insect, but not having a rank 

 smell, was liberated in the cage. It flew against the sparrow, fright- 

 ening him badly; but in five minutes the bird had recovered his cour- 

 age, and, giving chase to the butterfl}^, captured it, after several 

 minutes of lively fluttering, and finally succeeded in killing it and 

 eating some of the viscera. If the chase had occurred out of doors, 

 the butterfly would have had no difiiculty in getting away. Medium- 

 sized brown millers (Noctuidse) were eaten with great relish, despite 

 the fact that they were seldom found during stomach examinations, 

 which suggests the j)ossibility of their being overlooked because of 

 the diflicult}^ of recognizing their remains among the comminuted 

 contents of stomachs. It is not improbable that these insects are 

 frequentl}^ eaten bj^ some species of sparrows. 



Some experiments in feeding the song sparrow with difl'erent kinds 

 of seeds were attempted, but he did not lend himself with any degree 

 of enthusiasm to this line of work. Lamb's-quarters, amaranth, and 

 polygonums were frequently placed in the cage, but onl}^ starvation 

 could bring him to eat any of them, probably because he had been 

 supplied too long with canary seed, which seemed to suit his palate. 

 Experiments with dandelion and amaranth seeds have already been 

 referred to (see pp. 22 and 26). The seeds of chickweed and pigeon- 

 grass were eaten with great relish. It was interesting to note the way 

 in which the bird hunted for food of this kind. He searched about the 

 bottom of the cage, sometimes hopping, sometimes walking, and when 



