9 
collection from Pea Ridge, Arkansas. Swarth distinctly states that they 
do not migrate down through California, with which, after a careful exam- 
ination of the Red-tail series in the Museum of Vertebrate Zoology, the 
writer agrees. It, therefore, seems safe to say that Harlan’s Hawk breeds 
in northern British Columbia, southern Yukon territory, and adjacent 
parts of north Alaska, perhaps irregularly south to southeastern Alberta or 
even Dakota, migrates across the Canadian prairies to the Mississippi 
valley, and winters in the area between Arkansas and Missouri rivers. 
It is probably a small race numerically, with a narrow migrational route, 
and of local residential distribution. 
conclusions 
A summary of the foregoing distribution indicates that the Red-tailed 
Hawk of eastern Canada is pure borealis, a bird of practically constant 
character. The prairie provinces are inhabited by birds that are pre- 
dominantly borealis, but with calurus features, more or less common 
intrusion of krideri, and at least a sporadic influence of harlani towards 
the west. Southern British Columbia is inhabited by practically pure 
calurus, and the far northwest, at least centring about the British Columbia- 
Yukon- Alaska boundary intersection, contains birds that are predominantly 
harlani, but with strong intermixtures of calurus and krideri. 
Borealis has an extreme eastern distribution and calurus an extreme 
western one and they are the only forms that occur anywhere practically 
free from other strains. Krideri occurs along the western boundary of the 
borealis range and harlani at the northern extent of calurus territory, 
neither having any centre of distribution where they occur in purity. 
The conclusion seems obvious, that whereas borealis and calurus are good 
subspecies in the modern acceptation of the term, krideri and harlani are 
only colour phases, the former of borealis, the latter of calurus. 
The suggestion of geographic distribution exhibited by both krideri 
and harlani, especially the latter, does not necessarily deny the conclusion 
in regard to polychromatism. Nowhere do they appear in pure strain and 
we find similar local single-phased communities in acknowledged dichro- 
matic species, of which the Screech Owl and Grey Squirrel are outstanding 
examples. The geographic separation of dark and light-phased Swain- 
son’s Hawk is also a case in point. The black phase of this species is far 
more common to the north of the prairies than near the International 
Boundary and Major Brooks tells the writer that the black Swainson is 
characteristically a bird of the higher elevations in British Columbia, as 
the light one is of the valleys. The writer views harlani , in similar but 
more strongly marked case, as a local chromatic form of calurus, whereas 
krideri represents a dichromatism limited to the western representative of 
borealis, but transmitted to calurus at the northern extension of the latter’s 
range. 
There remains a nomenclatural difficulty regarding Buieo harlani 
Audubon ( Buteo borealis harlani of the Check-list). Harlani is based on 
Plate LXXXVI, of Birds of America (folio) 1830, with which the description, 
vol. I, page 40 (octavo), agrees perfectly. Incidentally the writer has also 
examined the original drawing for this plate in the Natural History Society 
of New York to interpret the intention of the author and correct the slight 
