198 
strictly personal experience. Is there no instinctive avoid 
ance here, as elsewhere, in consequence of the inheritance 
of ancestral tendencies? Is there no parental instruction, 
as in many cases there undoubtedly is? If so, disagree 
able insects would not be attacked at all. Moreover, in 
many cases insects would not necessarily suffer if they 
were attacked, because they would be saved, as Mr. Wallace 
has shown, by the eye-spots upon their wings, or, as 
Mr. Trimen testifies, by the extreme elasticity of their 
whole structure. 
But if we assume, for the sake of argument, that the 
resemblance as it now exists is useful to both species, 
this fact is far from proving that it has been produced 
by Natural Selection. For the idea is, that the fewer 
signs of inedibility, the better it will be for the foolish, 
inexperienced young birds, and still more for their victims. 
But if this is so, how confused these same young birds must 
have been when these two protected species were in the pro 
cess of growing like one another. Two modes of arriving 
at this result are conceivable. One species may adopt the 
livery of the other. In that case, you will have the models 
of one pattern, and the unmodified members of the 
mimicking species of another pattern, and between the 
two, one, two or several gradations not like either of 
them, and yet all of them nasty to eat. If both species 
undergo changes in the same direction, the confusion will 
be greater rather than less. How is Natural Selection 
to aid in a case like this ? 
But whether these objections have any weight in them 
or not, the fact remains that unprotected species mimic 
unprotected species. In this case of false (!) mimicry it 
is admitted that Natural Selection is necessarily excluded. 
Some other explanation must be found ; and I see no 
reason for doubting that in this, as in other cases, the