( xxxix ) 



the experiments left in me the irresistible conviction that 

 our birds at Chirinda are anything but specially reluctant to 

 prey on butterflies. And I know at present of no reason why 

 the birds of Chirinda should be different in this respect from 

 the birds of all those localities in which the evidence is as 

 scanty to-day as it was at Chirinda until I specially tested 

 the point. 



There were further lines of evidence too. Newly captured 

 birds, where tested, showed in their refusals and acceptances 

 a good previous knowledge of their butterflies; and damage 

 to butterflies' wings that was quite indistinguishable (as my 

 exhibit to-night most convincingly shows) from the common 

 form of damage one sometimes finds in nearly every 

 fairly high-grade butterfly one meets, and that occurs in 

 butterflies collected all over the world, was actually seen in- 

 flicted on a number of occasions by both wild birds and tame. 

 Moreover, on my painting conspicuous eyespots on the under 

 surface of more than fifty large butterflies {Charaxes) that 

 used to come to feed at the bananas in my verandah, out of 

 47 injuries subsequently sustained by the wings, 36 were 

 at or just beside eyespots, and 25 of these affected both 

 of a pair of wings. " Accidental damage " fails to fit the 

 case. Nor does it sufficiently meet the case (with which I 

 have many times met) in which abdomen and wing are corre- 

 spondingly wounded. It is hard, too, to believe that the 

 beautiful impressions of a bird's bill that occasionally accom- 

 pany or replace wing wounds in the butterflies we capture, 

 and that occurred too in the wet paint of some of the butter- 

 flies of the just-described experiment, have been placed there, 

 as used to be believed of fossils, merely to tempt us to 

 error ! 



With all this very convincing direct and indirect testimony 

 the striking lack of evidence hitherto afforded by stomach- 

 examination is very difficult to reconcile — at any rate, the 

 result of the great American examinations, in which the 

 microscope was, I imagine, freely used. Having regard to 

 relative population, we cannot expect to find more than 

 occasional butterflies in stomachs, but five out of 50,000 

 is not even " occasional." My own lack of evidence — and 



