WARNING COLOURS AND MIMICRY 159 



these non-mimetic races have suffered less hardship and 

 persecution than those which have gained the protection 

 of mimicry. In the case of the Madagascan race this view 

 is certainly plausible, and is adopted by Professor Poulton. 

 " It requires a very slight exercise of the imagination/' he 

 writes, " to picture the steps by which these marvellous 

 changes have been produced ; for here the new forms 

 have arisen at so recent a date that many of the inter- 

 mediate stages can still be seen, while the parent form has 

 been preserved unchanged in a friendly land, where the 

 keen struggle of continental areas is unknown." 



Let us now turn once more to Plate XXIII. The two 

 distasteful models which are mimicked by the Dismorphia 

 are practically identical in appearance, yet they belong to 

 distinct sub-families. We see, therefore, that in addition 

 to the mimcry of protected species by harmless ones, there 

 is a mimicry within the protected groups themselves. 

 Cases of this kind were remarked upon by Bates, but he 

 did not explain them. It remained for the late Dr. Fritz 

 Muller, a German naturalist resident in Brazil, to show 

 that if warning colours and mimetic resemblance are 

 really beneficial to insects, no limit can be set to the 

 sphere of their usefulness. He proved that it must be 

 advantageous for protected species to resemble each other 

 for the reason that these species are never absolutely 

 exempt from hostile attack. The usefulness of warning 

 coloration depends upon the fact, established by number- 

 less experiments, that young insectivorous creatures do 

 not inherit . a knowledge of what they may eat with 

 impunity, and what they ought to avoid ; they have to 

 learn by actual tests as they grow from youth to maturity ; 

 and this means that a certain number of victims must be 

 drawn each season from the ranks of every protected 

 species, no matter how glaringly conspicuous its warning 



