52 Mr. E. B. Poulton's notes upon the colours 



food-plant, and, having succeeded, has gone on to find 

 many more, will appreciate the great amount of truth 

 that there is in the popular notion of the eye "getting 

 in." The true explanation for this kind of work is, of 

 course, that we can never appreciate the true relation of 

 a larva (protected by colour, &c.) to its surroundings 

 until we have had actual experience of it. This is 

 especially true of the larvse which depend upon general 

 jDrotective resemblance. And, having become once 

 accustomed to the disguise in a particular instance (the 

 reality of which is shown by the difficulty in getting 

 accustomed to it) it is comparatively easy to detect other 

 similarly protected instances. The reason why descrip- 

 tion can never take the place of experience has been 

 shown in the discussion of "general protective resem- 

 blances" ; it is because the harmony is so subtle that it 

 cannot be understood without most careful observation 

 of the perfectly normal and undisturbed larva on its 

 food-plant. Even then, as was pointed out, although 

 the disguise may fail before an experienced observer, the 

 complex conditions which render it generally successful 

 — in fact the disguise itself — may not be understood in 

 the least. If this be true it is obviously an advantage 

 to a larva to appear under two forms possessing 

 respectively the colours which are most (generally) pro- 

 tective : an especial advantage if the colours are added 

 to a form with much special protective resemblance. 

 There is, in fact, a special protection for both forms, the 

 yellowish-green larvae resembling young green birch 

 twigs, the brown larvae resembling older twigs. 



It is very likely that an entomologist who had never 

 seen either form would continue finding the form which 

 he first discovered, and would fail to see the others (after 

 searching a tree twig by twig in the manner of the 

 enemies of the larvae). Or, if he found both forms, he 

 would find one more easily and frequently than the 

 other, that, namely, to which he was more accustomed, 

 and he could not become as accustomed to either as he 

 would have been to the larva if monomorphic. There 

 might often be exceptions to this, but if it ever happened 

 the species would gain by larval dimorphism. There is 

 every reason to believe that the natural enemies of the 

 larvae are similar to man in the respect above-mentioned. 



In other kinds of protection we argue from the effect 

 produced by certain colours, forms, or attitudes upon 



