218 THE ORIGIN AND DEVELOPMENT 



butterflies, that nevertheless these colors are often 

 very pliable ; for instance white, yellow, and 

 orange are readily interchangeable ; and a similar 

 reciprocal relation exists between orange and red, 

 or between certain tones of yellow and brown. 



Again, when we compare individuals of the 

 same species, especially if of opposite sexes, the 

 great difference between them in the profusion 

 with which scales of metallic blue are sprinkled 

 upon a normal dark ground shows how easily 

 black or dark brown is replaced by metallic 

 blue ; the latter again is readily interchange- 

 able with metallic green, or with purple or or- 

 dinary blue or green. Some instances of this 

 interchangeability must be given farther on, and 

 it is not therefore worth while to dilate upon it 

 here. It should, however, be stated that the iri- 

 descence on the wings of many butterflies is of an 

 entirely different nature, being due to a micro- 

 scopic striation of the outer surface of the scales. 



The hypothesis, then, of the origin and devel- 

 opment of color-patterns in butterflies which I 

 would maintain is based upon the scale of compli- 

 cation seen in the markings of living Lepidoptera, 

 and the prevalence among them of certain special 

 patterns. According to it, the wings first showed 

 signs of divergence from uniformity by a deepen- 

 ing of the color next the outer margin, which 

 thereafter became separated into distinct trans- 



