108 University of California Publications in Zoology (Vou. 30

coloration seen in calurus is described as young harlani, a mistake
that is pointed out in Baird, Brewer and Ridgway (1874, vol. 3,
p. 294).

At first glance it seems startling to ascribe to the Harlan hawk a
far northern breeding habitat. In the A. O. U. Check-list (1910,
p. 158) the range given is as follows: ‘‘Lower Mississippi Valley and
Gulf States, from Louisiana to Georgia and Florida; casual in Colo-
rado, Texas, Kansas, Nebraska, Iowa, Illinois, and Pennsylvania.’’
I cannot find, though, that there are definite published accounts of the
breeding of harlani in any region whatever. Audubon’s belief that
the birds he shot near St. Francisville, Louisiana, had bred in that
vicinity was based on hearsay. He shot his birds in November (see
Coues, 1880, pp. 202-203) and had no first-hand knowledge of their
nesting. Beyer, Allison and Kopman (1908, p. 442) in their ‘‘List of
the Birds of Louisiana’’ state: ‘‘None of the writers has evidence of
its breeding in Louisiana.’’ Jt seems to me, in the absence of any
positive published statements, that the assumption that the breeding
ground of the Harlan hawk is in the Gulf states is an utter mistake.

Besides the Atlin series there are at hand three specimens of hawks
from the northwest that I think are referable to harlami. These are
two young birds (nestlings), from a point sixty miles below Forty-
mile, Yukon Territory, July 28, 1894, collected by C. L. Hall ( Mus.
Vert. Zool. nos. 4966, 4967) ; and an immature male (Mus. Vert. Zool.
no. 42048), a migrant, shot by the present writer in Kispiox Valley,
near Hazelton, British Columbia, August 27, 1921 (see Swarth, 1924,
p. 336).

These birds in life were extremely puzzling. While there was
much to suggest Buteo borealis in the actions of the living bird, the
uniformly dark coloration brought B. swainsoni to mind, and an occa-
sional glimpse of white marked rectrices in a bird wheeling in distant
flight was distinctly suggestive of Archibuteo. With specimens in
hand, Buteo swainsont and Archibuteo were quickly eliminated, of
course, but other questions remained.

The status of the Harlan hawk as a distinct subspecies has been
questioned. Our own findings in the Atlin region, while not assumed
to be a final disposal of all the difficulties involved, do seem to place
this form in a more secure position as a geographic race than it has
yet enjoyed. The birds were abundant and nesting over a wide
expanse of territory, and within that region they were the only form
of Buteo borealis that was seen. Parents and young were seen together