( XXXV ) 



of bark and scrutinise the underlying surface, or has seen 

 the closeness with which birds will sometimes examine an 

 insect that they are hesitating to attack or (occasionally) 

 one that they have just rejected, or the way in which 

 some birds will try inanimate objects that they are not 

 quite sure about, will continue to retain an implicit trust 

 in the argument of " Hypertely." And no one who has 

 frankly noted and analysed, day after day, his own hesita- 

 tions and mistakes (many of them quite ridiculous), whether 

 due to an imperfect view, the play of light and shadow, 

 lack of recent acquaintance, insufficiently concentrated 

 attention or attention concentrated too late — and a mere 

 splash of colour or trick of flight in common — and has seen 

 (as I have) similar hesitations on the part of birds, will doubt 

 the value — incomplete, but very real — of incipient resemblances. 

 As for instinctive knowledge of food-values, special experiments 

 on several of my birds have produced the same results as 

 those of Prof. Lloyd Morgan, and I have seen even wild birds 

 test and reject inanimate objects. A special experiment on 

 a young eagle {Aquila wahlbergi) showed that, unstarved, it 

 would eat vegetable substances (including, once, raw green 

 peas and a tuberous PlectrantJms resembling Jerusalem 

 artichoke) as readily as meat, even picking them up from the 

 ground for itself, until (as seemed to be rather clearly shown) 

 it learnt which food was, very definitely, disagreeing with it. 



The eating of butterflies by birds. — I have myself seen far 

 more numerous attacks on insects of other orders, and 

 watched dipteron-eaters that were being kept busy by their 

 favourite prey consistently ignoring butterflies; and in my 

 days of superficial stomach-examination I found butterfly 

 remains in only five stomachs out of more than a thousand. 

 American stomach-examinations are said to have produced 

 only five cases out of 50,000, I am therefore unable to 

 regard it as a matter for surprise that those to whom mimicry 

 does not appeal should have been sceptical about the eating 

 of butterflies by birds. 



" Neglect of well-directed and sustained observation " (Mr. 

 Trimen), frequent removal of wings and the likelihood " that 

 attacks will in general be made only under specially favour- 



