54 MR. J. C. MOTTKAM ON THE 



(5) The animals appreciate a difference in palatability. Possibly there 

 are other factors which control the selection. It is obvious that in order to 

 estimate the palatability of. food all these determining factors must be taken 

 into account. 



It may here be mentioned that relative palatability may itself depend upon 

 several factors, of which taste and digestibility are not necessarily the most 

 important. The readiness with which the prey can be disposed of is an 

 important one : for instance, when butterflies are attacked by small birds the 

 large expanse of wing protects the small body, the insect is likely to be 

 seized by one or both wings, when the bird shifts its hold in order to dispose 

 of it the wing is momentarily released and the scarcely injured insect 

 escapes, requiring to be recaptured ; further, when the insect has been 

 killed, the stiff dry scale-covered wings are difficult to swallow, and are, as a 

 rule, removed before the body is eaten. It would appear that a butterfly 

 is unpalatable in this respect, apart from taste. If warning coloration 

 advertises unpleasant taste there seems no reason why it should not also 

 advertise difficult disposal. 



It is also conceivable that warning coloration may similarly advertise 

 difficult capture ; on the other hand, conspicuous coloration would give the 

 hunter a good mark to follow. Inconspicuous coloration is of use during 

 motion: a covey of grouse flying over heather can only be followed with the 

 eye for a short distance ; should there be a partially white bird among the 

 covey it can be followed over the moors for a mile or so. Entomologists 

 find some inconspicuous butterflies very hard to capture on the wing, not 

 because of rapid flight, but because they are being constantly lost in the 

 background. 



The feeding-habits of birds upon butterflies is of special interest because 

 selection by birds has been considered to be a factor in the production of 

 mimicry in these insects. On this account, particular attention has been 

 paid to the feeding of birds upon butterflies. Marshall (1) has collected 

 together the records of the attack of birds upon butterflies up to the year 

 1909 and given reasons for the paucity of the evidence, which has been 

 considerably increased since then. The evidence at present collected appears 

 to show that — 



(1) Birds seldom feed upon butterflies ; 



(2) They are less often attacked, and are less often found in the alimentary 

 canal of birds, than are many other insects. 



(Further evidence may, however, negative both these statements. When- 

 ever accurate field-observations are made these instances of birds feeding 

 upon butterflies are by no means infrequently recorded. Swynnerton has 

 shewn that these insects are so broken up in the alimentary tract of birds 

 that a careful microscopical examination is required for their recognition.) 



