Notes in 1887 upon lepidopterous larve, de. 585 
in the webs of spiders: hence it appears that they keep 
out of the way of these animals. And we have some grounds 
for the belief that the additional refinements may not be 
entirely lost upon the insect enemies. At any rate, I have 
shown that if we are to accept the theory of sexual selection 
(and no other theory has yet been suggested which can serve 
- as an adequate explanation of all the facts of the case), 
the comparison between the colours and patterns which 
are due to this principle, and those which are due to the 
various defensive needs of the insect, leads us to the 
inevitable conclusion that ‘four own sense of what is 
beautiful, entirely comcides with that of an insect” (See 
Proc. Zool. Soc. Lond., March, 1887, p. 216). I should 
not, however, have ventured to speak so plainly of the 
meaning of the various details in the defensive attitude 
of the larva, if I had not been able to rely upon the best 
support attainable—the support yielded by direct experi- 
ment. 
In the first place, my interpretations were obtained 
from the careful study of the living larva—at rest and 
after various degrees of irritation. The terrifying atti- 
tude is often imperfectly assumed at first, and the full 
meaning of the details is only seen after continued 
irritation ; but at the same time this very fact carries 
conviction with it. The larva at first relies on the effect 
of its outstretched quivering legs, and the posterior part 
of the body is not usually completely doubled over. As 
the irritation is repeated and increased, the larva 
gradually adds the various details which go to make up 
the terrifying attitude in its most perfect and elaborate 
form. 
In the second place, I offered the larva to two of its 
vertebrate enemies, and actually witnessed the effect 
produced upon the latter. The larva was placed upon a 
table, and was made to assume the terrifying attitude in 
a very complete manner. The attention of a marmoset 
was then directed to the larva, and although the former 
is excessively fond of insects, and seizes caterpillars with 
greatest avidity, it was much impressed by the alarming 
sight, and only ventured to attack the larva after the 
most careful examination, and even then in the most 
cautious manner. Meeting with no resistance, the larva 
was soon devoured and greatly relished. I then tried a 
similar experiment with a lizard, and the same results 
