6 MY LIFE [Chap. 



can understand that to get the full benefit of this protection 

 they should be easily recognized, should have some outward 

 character by which birds would soon learn to know them and 

 thus let them alone ; because if birds could not tell the eat- 

 able from the uneatable till they had seized and tasted them, 

 the protection would be of no avail, a growing caterpillar 

 being so delicate that a wound is certain death. If, therefore, 

 the eatable caterpillars derive a partial protection from their 

 obscure and imitative colouring, then we can understand that 

 it would be an advantage to the uneatable kinds to be 

 well distinguished from them by bright and conspicuous 

 colours. 



" I may add that this question has an important bearing 

 on the whole theory of the origin of the colours of animals, 

 and especially of insects. I hope many of your readers may 

 be thereby induced to make such observations as I have 

 indicated, and if they will kindly send me their notes at the 

 end of the summer, or earlier, I will undertake to compare 

 and tabulate the whole, and to make known the results, 

 whether they confirm or refute the theory here indicated. 



" Alfred R. Wallace. 



"9, St. Mark's Crescent, Regent's Park, N.W., 

 " March, 1867." 



This letter brought me only one reply, from a gentleman 

 in Cumberland, who informed me that the common "goose- 

 berry " caterpillar, which is the lava of the magpie moth 

 {Abraxas grossulariata), is refused by young pheasants, par- 

 tridges, and wild ducks, as well as sparrows and finches, and 

 that all birds to whom he offered it rejected it with evident 

 dread and abhorrence. But in 1869 two entomologists, Mr. 

 Jenner Weir and Mr. A. G. Butler, gave an account of their 

 two seasons' experiments and observations with several of 

 our most gaily-coloured caterpillars, and with a considerable 

 variety of birds, and also with lizards, frogs, and spiders, con- 

 firming my explanation in a most remarkable manner. An 

 account of these experiments is given in the second and all 

 later editions of my book on " Natural Selection ; " but it 



