216 THE ORIGIN AND DEVELOPMENT 



If, on the other hand, the break be supposed to 

 occur at the veins themselves, then the tendency 

 would be to form short transverse bars, or quadrate 

 or more or less rounded spots in the interspaces ; 

 and, finally, by a differentiation of the exterior 

 and interior portions of a round spot, a more 

 or less perfect ocellus would be formed. Occa- 

 sionally we find long streaks of dark color down 

 the middle of the interspaces, similar to those 

 along the veins, produced, no doubt, by the fre- 

 quent presence of a crease in such places, and the 

 tendency of scales to follow it ; the comparative 

 weakness of such breaks in the continuity of the 

 membrane is the reason of the comparative rarity 

 of this form of ornamentation. The formation of 

 ocelli has been shown by Darwin, who traced, in 

 specimens of a South African butterfly (Cyllo 

 Leda) a perfect gradation " from excessively mi- 

 nute white dots, surrounded by a scarcely visible 

 black line, into perfectly symmetrical and large 

 ocelli ;" and instances are common in our own 

 butterflies where one can follow a similar series 

 onward from a uniform circular dark spot. First, 

 a central white dot appears in it ; next the whole 

 is encircled by a light-colored halo, and so on. 

 Darwin mentions one moth with a magnificent 

 ocellus consisting of a black centre with eight 

 concentric zones of colors. 



Ocelli not infrequently surpass the bounds of 



