690 Mr. R. C. L. Perkins 07i the 



cult to understand how two species of these di&tinct genera 

 can under totally different conditions of cHmate and en- 

 vironment produce remarkable colour varieties, totally 

 dissimilar from their usual forms, yet almost identical 

 with each other, unless they be reversions to a former 

 style of coloration. 



[No further quotations from the Introduction will be 

 founil beyond this point, but it has been necessary in the 

 preceding paragraphs to quote from it somewhat exten- 

 sively, in order that the discussion in the following letters 

 may become clear. After reading the statements repro- 

 duced above, I asked Dr. Perkins, among other questions 

 bearing on a possible Miillerian interpretation of the facts, 

 whether the reversion to an ancestral pattern — or more 

 probably the persistence of an ancestral pattern — in the 

 form distinctus, might not be associated with the presence 

 of the pale-banded Odyneri wliich are also found in the open 

 country. He replied, Nov. 15, 1911, as follows: — ] 



X. notostidus, the hlack-hodiedL form of distinctus, seems 

 to be the only form in the forest region where are no pale 

 banded Odynerus, except occasional reversional individuals. 

 Typical distinctais of Smith is essentially an open country, 

 sublittoral form, but the notostidus form may occur with 

 it, and intermediates. There is a number of pale-banded 

 Odynerus, belonging to this open country, or sublittoral, 

 and only belonging to this coimtry. Several species of the 

 predominant black group of Odynerus are common both in 

 this open country and forest alike. This would be very 

 suggestive to the Miillerian. 



The case of var. bidecoratus is quite different, for instead 

 of being coastal, it inhabits very wet forest districts, mixed 

 with the typical form but rarer, and probably less widely 

 distributed. Before I knew this, I thought the pale 

 marked Crabronid vars. migiit be produced by the dryness 

 and heat of the coast region — they average smaller in size 

 also : bidecoratus upsets this view. 



Mtillerians would say that ' notostictus ' persisted in the 

 coastal regions because of the presence of the pale-banded 

 Odynerus (or, at least, for the same reason that the latter 

 do, viz. absence of enemies), and would cite the fact that 

 all Crabronids on Kauai are yellow-banded, the black- 

 bodied group of Odynerus being absent there. Obviously 

 the colour of the var. bidecoratiis is quite out of place in 

 wet forests on Hawai, where are no yellow-banded Odynerus, 



