152 MIMICEY IN N. AMERICAN BUTTERFLIES 



THE DANAINE MODELS OF NORTH AMERICA, 



AND THEIR RELATIONSHIP TO THE SOUTH 



AMERICAN AND OLD WORLD DANAINAE 1 



The Danainae are the most important and most 

 extensively mimicked of all specially protected 

 butterflies in the Old World tropics. The Acraei- 

 nae, so abundant in Africa, are also greatly 

 mimicked, but to a far less extent than the com- 

 paratively few species of Danainae found in the 

 same Region, all belonging to the section Danaini. 

 The Ethiopian Acraeas in fact supply several 

 mimics of the Danaines, but no example of the 

 opposite relationship is known. In the tropical 

 East, the Acraeinae are poorly represented, while 

 the Danainae (Danaini, Euploeini, Hestia, Hama- 

 dryas) are dominant in numbers as well as in the 

 power of influencing the patterns of other butter- 

 fly groups. In both Africa and the East, Miil- 



1 The subject of the address from this point onwards is treated 

 in considerable detail in the author's memoir, Mimetic North 

 American species of the Genus Limenitis (. /.) and their models, 

 in Trans. Ent. Soc. Lond., 1908, 447-88. Dr. Jordan's later con- 

 clusions as to the affinities of Danaida plexippus, added to the 

 memoir in a terminal note (488) and somewhat at variance with 

 his earlier conclusions quoted in the text, are here adopted 

 throughout. A broader and less detailed treatment is followed 

 in this address, special attention being directed to the numerous 

 points on which further observations are required. Where no 

 other authority is mentioned I have followed the synonymy and 

 geographical distribution of Scudder's great work, Butterflies of 

 the Eastern United States and Canada, and, for the Papilionidae, 

 Rothschild and Jordan's fine monograph (Nov. Zool., xiii, 1906, 

 411-752). I have not, however, followed Scudder in the general 

 use of Basilarchia as a generic name, because I think that the 

 whole group of Limenitis, in its widest acceptation, requires revision, 

 and that until this has been accomplished it is inexpedient to 

 adopt the terminology proposed for a portion of it. 



