( Ixii ) 



and the mixture of groups was probably to be explained by 

 accidental transport from one island to another. 



This was, in fact, Dr. Perkins's interpretation of the 

 existence of two Colour-groups on the most isolated of all the 

 islands, Kauai ; for he remarks that " excepting two species 

 (. . . probably recent derivations from similar forms on other 

 islands) the Kauai wasps have become superficially all alike." 

 Such complications are of course far more likely to occur in 

 the central islands of the chain — nearer together and liable to 

 receive immigrants from both directions. 



The following was an attempt to reconstruct the histoi-y of 

 the Colour-groups within the islands. It, in the main, followed 

 Dr. Perkins's account, but included a few suggestions bearing 

 on the mixture of the groups. 



(1) The original yellow-banded pattern persisted at any rate 

 in Oahu (the island nearest to Kauai), and probably through- 

 out the islands, until after the Structure-groups had been 

 formed and Kauai had received the immigrants which have 

 produced its dominant banded Coloiir-group II (= A). 



(2) The black Group I (= E) then arose in Hawaii, perhaps 

 in consequence of the arrival from Asia of the second immi- 

 grant ancestor, 0. nigrijieimis. On the other hand, in the 

 specimens sent by Dr. Perkins, the wings of some of the 

 Odyneri were so much darker and more iridescent than those 

 of the nigripennis as to throw some doubt upon the hypothesis 

 that the latter had acted as the model. After the group was 

 formed, Hawaii became a centre for the occasional accidental 

 dispersal of black species to Maui and further north-westwards 

 to other islands ; or the spread of Group I may have followed 

 the dispersal of the black-bodied, dark-winged 0. nigri])ennis, 

 which Dr. Perkins described as the most dominant and wide- 

 spread species on all the islands except Kauai. That the 

 black Group I is oldest on Hawaii was indicated by its almost 

 complete dominance in that island. 



(3) On Kauai, 0. nigripennis became absorbed into the 

 domin.ant Colour-group, giving rise to the yellow-banded 

 0. radida, F. This species is as abundant on Kauai as 

 nigripennis on the other islands, and Dr. Perkins suggests the 

 possibility (among others) of a diaposeniatic relationship, the 



