168 MIMICRY IN N. AMERICAN BUTTERFLIES 



costal end of the white band is retained, often 

 for the full breadth of the marking, forming 

 a linear streak. 



I have dwelt upon the changes undergone by 

 the white band as an example of the way in which 

 the new markings have been carved out of the 

 old. The changes in the elaborate marginal 

 pattern would have been equally convincing as 

 evidence for a gradual and * continuous ' trans- 

 formation. 



THE MODIFICATION OF THE LIMENITIS MIMIC 

 OF PANAIDA PLEXIPPUS INTO A MIMIC OF 

 D. BERENICE IN FLORIDA 



Danaida plexippus occurs together with D. bere- 

 nice in Florida, but the latter far outnumbers the 

 former, and the modification of Limenitis archippus 

 into the form floridensis, Strecker (= eras, Edw.) 

 is probably entirely due to the predominance of 

 one model over the other. Data for determining 

 the exact proportions in various localities would 

 be of high interest. There is no reason for 

 believing that Berenice is in any way more or less 

 distasteful than plexippus, but its abundance makes 

 it a more conspicuous feature in the environment. 



It is evident that the change has been of the 

 kind expressed in the above heading ; for, as has 

 been already implied on pp. 162-3, traces of the 

 former Mimicry of plexippus persist in floridensis 

 and tend to detract from the resemblance more 



