112 USEFUL BIRDS. 



caterpillars {Leucarctia acrmiC), the caterpillar of the vaporer moth 

 ( Orgyia) and the spiny larva3 of buttertiies ; with these perhaps may 

 be classed the European cm-rant sawfly. He was disposed to consider 

 the "flavor of all these caterpillars as nauseous, and not that the 

 mechanical troublesoraeness of the hairs prevents their being eaten. 

 Larvae which spin webs, and are gregarious, are eaten by birds, but not 

 with avidity ; they appear very much to dislike the web sticking to their 

 beaks, and those completely concealed in the web are left unmolested. 

 When branches covered with the web of Hypononienta evonymella (a 

 little moth of the Tinea family) were introduced into the aviaiy, those 

 larvae only which ventured beyond the i^rotection of the web were eaten." 

 " Smooth-skinned, gaily colored caterpillars (such as the currant Abraxas 

 or spanworm), which never conceal themselves, l^ut, on the contrary, 

 appear to court observation," were not touched by the birds. He states, 

 on the other hand, that "all caterpillars whose habits are nocturnal, 

 and are dull colored, with fleshy bodies and smooth skins, are eaten with 

 the greatest avidity. Every species of green caterpillar is also much 

 relished. All Geometni?, Avhose larvae resemble twigs, as they stand 

 out from the plant on their anal prolegs, are invariably eaten." * 



Such statements as these are at least interesting, but they 

 must be classed as negative evidence, and cannot justify the 

 assertions so often made that birds do not eat hairy cater- 

 pillars, when there is convincing, positive evidence that cer- 

 tain species do eat them. This statement that birds do not 

 eat such caterpillars, which has been so long reiterated, 

 parrot-like, by one Avriter after another, is entirely at variance 

 with my experience, and my opportunities for investigating 

 this subject probably have been better than those of most 

 observers. The great burden of proof is upon those who 

 make the allegation, for it is always hard to prove such 

 sweeping generalizations, and often not at all difficult to dis- 

 prove them. A naturalist may with propriety say what he 

 has seen a bird do, but he should be cautious in stating what 

 it does not do. The reiterated assertion that hairy cater- 

 pillars are immune fi'om the attacks of birds has been modi- 

 fied of late by some writers, and is now oftener given, in 

 effect, that few birds eat them ; but this statement needs still 

 further modification. We cannot rely on results secured by 



1 First Report on Injurious aud Beneficial Insects of Massacliusetts, l)y A. S. 

 Packard. Annual Report of tlie Massachusetts State Board of Agriculture, 

 1870-71, pp. 358, 359. 



