36 



Accofdino: to the theovy of mimicry as due to the influence of 

 environment, we should expect that the Old World invaders when 

 they entered North America, would have acquired a North American 

 pattern. We find, however, that they retain the characteristic 

 appearance of an Old World group of Danaines, and that they 

 impress this pattern upon the ancient North American genus 

 Limcnitu, producing admirable mimics. The facts are, therefore, 

 the reverse of what Ave should expect on the theory of mimicry as a 

 direct result of geographical conditions. 



If the mimetic pattern of Lliinniith arc/ti/ipiifi be compared with 

 that of the ancestral forms of the genus, it will be seen that those 

 parts of the old markings that were available for the purpose, have 

 been retained and worked up into the mimetic resemblance, while 

 other parts that would interfere have been removed. A single 

 element in the old pattern which interferes with the resemblance 

 has been retained, namely, the black line crossing both the upper 

 and lower surfaces of the hindwing. This line is clearly the black 

 outer border of the old white band of the genus, and traces of white 

 may still be found clinging to its inner border on the under surface 

 of many specimens. It is extremely interesting to find that in the 

 Hudson River district the black line has disappeared from the upper 

 surface of the hindwing in a large proportion of the males, while 

 others show a transition. 



In Florida the model D. ple.rippus exists side by side with D. 

 berenice, but the latter is far commoner, and influences L. arrhij)pu>> 

 more strong!}', so that the mimic here gains the rich chestnut-brown 

 Goloux oi berenice, and is known as the local ionn Jioriilemi>i {eras). 

 There can belittle doubt that //()r/(/(^».s/,s is merely a geographical 

 race of areJiippua, and it would be extremely interesting to ascer- 

 tain whether the two combine to form an interbreeding com- 

 munity along the line of their overlap. That /ioridensia has been 

 produced by the transformation of cnrhippiis is proved by the fact 

 that the old archippns pattern, with its heavily marked veins, 

 can still be detected beneath the rich deep ground-colour, although 

 it tends to interfere with the resemblance to the new model berenice. 

 The blackened veins of the upper surface are to be looked upon as a 

 persistent vestige of an older mimetic resemblance. 



In Arizona we meet with the still more interesting mimic 

 Liiiienitis htihti, or obxoleta, resembling the local Danaine, ]>. 

 strit/nsa. I have recently had an opportunity for studying a long 

 series of this mimic, and find that its pattern is certainly more 

 ancestral than that of an-hipjiKu ^nd ihnidcnsis. It retains persistent 

 vestiges of the old white band on both surfaces of the hindwing, 

 and more evident traces of the same feature are retained by f he 

 forewing than in the other two forms. The most interesting vestige 

 is a short white stripe which represents the inner marginal 

 termination of the old white band on the upper surface of the 

 forewing. 



