268 THEORIES OF MIMICRY 



enormous number of facts observed in many very 

 different parts of the world. I believe that the generali- 

 zations will be admitted to be sound and to be well 

 warranted by the facts. Under any theory which is not 

 based upon selection, the whole of the facts on which 

 the generalizations rest become mere coincidences and 

 receive no explanation of any kind. Under Natural 

 Selection this vast body of facts becomes at once in- 

 telligible. Here the accumulated facts of the most 

 diverse kind, which receive an intelligible explanation 

 by the theory in question, yield a firm support to the 

 theory. There are many theories which are held upon 

 indirect evidence of precisely the same nature. We 

 believe in evolution, not because we see it taking place, 

 but because of the immense number of observed facts 

 which it renders intelligible. 



In the case of Natural Selection in relation to Mimicry 

 and Common Warning Colours it is to be confidently 

 hoped that direct evidence may yet be added ; indeed, 

 a considerable amount is even now forthcoming. Pro- 

 fessor Lloyd Morgan's work l upon the activities and 

 instincts of young birds of many species, proves that 

 their education is actually of the kind which is pre- 

 supposed in the theories of H. W. Bates and Fritz 

 Miiller. He shows that they have no instinctive know- 

 ledge of things which are good for food, but that they exam- 

 ine and test everything. On the other hand, they have 

 excellent memories, and retain a firm impression of the 

 appearance of objects which have given them an un- 

 pleasant experience. Furthermore, there was evidence 

 that they are influenced in their behaviour towards 

 other objects resembling the one which has proved ob- 

 jectionable to them. As to the aggressive Hymeno- 

 ptera, the evidence of their special methods of defence 

 is obvious to every one. With regard to specially pro- 

 tected groups of butterflies there is a large amount of 

 evidence from observation and experiment, but more is 

 to be desired. For positive results I may refer to Guy 



1 Habit and Instinct, London, 1896. 



