ATTACKS BY BIRDS 



123 



enemy which caused the injury. The illustrations form part of Marshall and Poul ton's 

 extensive paper on the bionomics of South African insects already referred to. I have here 

 endeavoured to give a condensed account of the large number of instances which have been 

 collected bearing on the present subject. No one who has studied Marshall's paper could 

 remain longer in doubt that birds are potent factors in the selective forces which guide and 

 control the evolution of butterflies. Their attacks are not exceptional phenomena, but 

 matters of habit and everyday occurrence. The manner in which such observation may 

 for long go unobserved is instanced by the case of the kestrel, which is now known to habitually 

 eat butterflies. If cases of this kind can remain so long unobserved in our own country, 

 where every wild creature has been carefully studied, it would indeed be rash to assert 

 that birds in tropical lands do not attack butterflies. It will probably be admitted that we 

 are not even acquainted with all the species of birds which inhabit the tropics, and our 

 knowledge of the economy of those species which have been described cannot yet be so 

 perfect as to enable any one to decide definitely as to their choice of prey. To sum up the 

 whole matter, there is on the one hand no sort of proof that birds do not exercise a selective 

 influence on butterflies, and hence that they 'have been but little concerned in the matter',^ 

 whilst on the other hand, where evidence to the contrary has been sought it has been found 

 in plenty, not only in reference to the preliminary contention, but still further in direct 

 confirmation of the hypothesis that some butterflies are less liable to attack than are others. 

 The mainstay of the argument of those who see so much weakness in the mimetic theory 

 suffers from the very form of instability which they allege in the contentions of its supporters — 

 it is based largely on assumption. 



As regards any early instinctive knowledge in birds of what to eat and what to avoid 

 there can be but little doubt that young birds learn only by experience to make a choice 

 of food. Those who have read Lloyd Morgan's work on ' Habit and Instinct ' will remember 

 that his young birds at first struck at anything of suitable size, whether eatable or not. The 

 list of objects which the author gives as forming their early prey includes chopped wax matches, 

 bits of paper, buttons, beads, cigarette ash, their own toes and those of their companions, 

 specks on the floor, and, in short, ' anj^thing and everything, not too large, that can or cannot 

 be seized is pecked at and, if possible, tested in the bill.' It is pointed out that there seems 

 to be no congenital discrimination between nutritious and innutritions objects, or between 

 those which are nice and those which are nast}^ The larvae of the Cinnabar moth were 

 tested and rejected, and the birds, after being fed with other larvae, still remembered the 

 former and refused them. Similarly, moorhen chicks soon discriminated between small 

 edible beetles and ' soldier beetles ', though, as the writer states, ' such discrimination is 

 not congenital, but acquired.' Again, in another passage, ' My observations with regard to 

 the effects of giving to young birds distasteful insects were made solely with a view to ascertain 

 whether there was any instinctive avoidance on the part of the birds. Of this I have not 

 found a single instance.' The author is careful to point out that there may be such instinctive 

 avoidance in young birds which, when hatched, are devoid of parental guidance, since ' such 

 guidance would in some degree prevent the incidence of natural selection, and would diminish 

 the elimination of those which without such guidance might eat to their destruction ' . * 

 Much more information on the same points is to be found in the work in question, which 

 I commend to the study of those interested in these matters. For the present purpose it 

 suffices to call attention to the fact that experimental proof is not wanting in regard to the 

 necessity of young birds learning what to eat and what to avoid. 



1 Quoted from Dr. Sharp's remarks on mimicry in ' The Cambridge Natural History '. 



2 It is not however intended by supporters of the mimetic theories that distasteful insects are also poisonous. 



Q 2 



