( xh ) 
forms with special protection and warning colours we should 
expect to find and did find attacked by certain special enemies 
able to disregard the means of defence and so gain for them- 
selves a supply of food which was abundant, easily seen, and 
easily caught. We should expect to witness such attacks 
more readily than any others, because the prey were them- 
selves slow flying, and locally abundant. Such facts were well 
known among the insects specially defended by stings, no less 
than in those protected by an unpleasant taste or smell. 
Thus bees were well known to be attacked by special birds, and 
a similar relationship to enemies would no doubt be found 
in all insects, however well defended. The same argument 
held with regard to procryptic colouring. It was erroneous 
to suppose that concealment was always efficacious; on the 
contrary, large numbers of insect-eating vertebrates preyed 
habitually on insects with procryptic colouring. Concealment 
was none the less an essential aid by which a species was able 
to keep up its average numbers because a vast number of 
individuals above the average were destroyed by enemies 
of one kind or another. The same facts held for the specially 
protected species with warning colours, only here we must 
look to the frequent attacks of a few specialised enemies as 
well as to a relatively enormous slaughter by parasitic insects, 
rather than to less frequent attacks by a very much larger 
number of enemies and a smaller amount of destruction by 
parasites, as in the procryptic species. One observation of 
Mr. Fryer’s seemed to him particularly significant—the 
Papilio agamemnon, L., which he had seen to be attacked and 
mutilated by the drongo. This observation fell into line with 
many other records of injuries seen to be similarly caused; and 
the whole body of such evidence might be fairly brought forward 
as proof that injuries of the same kind were inflicted in the 
sameway. Ifthis be admitted we were at once confronted by 
an immense body of circumstantial evidence indirectly proving 
the continual persecution of butterflies by birds. Collectors 
had only to look out for such evidence in the specimens they 
would generally be inclined to reject, in order to place beyond 
the reach of doubt the conclusion that butterflies are constantly 
attacked in this manner. 
