422 University of California Publications in Zoology. (VoL. 
To review the faunal characteristics as indicated by the fore- 
going lists, the Prince William Sound district possesses thirteen 
breeding land-birds in common with the Sitkan district (list 
no. 6), and nine in common with the Yukon district (list no. 5). 
Besides these, there are ten species which oceur throughout all 
three districts (list no. 4), and six which are not found in 
either the Yukon or Sitkan district (list no. 7). Of this last 
category, Canachites canadensis atratus is closely similar to 
C. c. osgoodi of the Yukon district; and as this bird has no 
analogue in the Sitkan district, it is clearly a recent derivative 
from the Yukon fauna. The same is evident of Lagopus rupes- 
tris kelloggac. On the other hand, Penthestes rufescens vivax 1s 
most nearly related to the P. r. rufescens of the Sitkan district, 
and has no near relative in the Yukon district. This subspecies 
may, therefore, be considered a contribution from the Sitkan 
fauna. However, as I have elsewhere suggested (Auk XNXI, 
July, 1904, p. 368), P. r. rufescens was itself, though much more 
remotely, derived from -P. hudsonicus of wide boreal range. 
Melospiza melodia kenaiensis, Passerella iliaca sinuosa, and 
Hylocichla guttata guttata are likewise most nearly allied to 
representatives in the Sitkan district; the same or near related 
forms extend also to the westward through the Cook Inlet region. 
That there are not more endemic species of birds in the 
Prince William Sound region, and that the species which are 
peculiar are so slightly differentiated, appear to be due to its 
small area, and to the relative narrowness and ineffectiveness of 
the barriers which surround it. The barrier for the Hudsonian 
species is the aretie-zone divide which surrounds the Sound in the 
form of a semicirele, with only a long, narrow coastal connectant 
alone the Kenai Peninsula to the westward, and another towards 
Yakutat Bay to the eastward. The only endemic arctic-zone 
species (Lagopus rupestris kelloggae) owes the opportunity for 
its differentiation to the isolation afforded by the mountains on 
the more remote islands of the Sound. (See p. 384.) 
That the Prince William Sound region has not produced as 
divergent forms of birds as the other faunal areas of Alaska 
may be due also to its more recent service as a center of differen- 
tiation. We may assume that there were no land-birds at all in 
