109 



variations in the proportions of the primaries and of the wings and 

 tail, in specimens of the same variety, was much greater than I had 

 expected to find ; in B. 'borealis^ for instance, the fourth quill is gener- 

 ally the longest ; but sometimes the fourth and fifth in B. montanus^ 

 generally the fourth in Nos. 5,836 and 19,925 ; the fifth in Ko. 4,545, 

 the third, fourth, and fifth, and in one, not in the table, the fourth and 

 fifth ; in calurus, in two specimens, the fourth, and in No. 16,026, the 

 fourth and fifth ; in Harlani, generally the third, but in No. 10, the 

 fourth; in insignatus generally the third, but in No. 6,955, the fourth ; 

 in Sivcdnsonii, generally the third, but in No. 8,539, the fourth, and in 

 No 8,540, the third and fourth ; in Bairdii generally the third, but in 

 Nos. 10,761 and 19,121, the third and fourth are the longest. The 

 same range of difierence will be seen to exist in the proportions of 

 the other primaries. The variation in number and shape of the 

 tarsal scales is considerable, as is usual in birds of this order. The 

 development of the festoon of the lower edge of the upper mandible, 

 one of the principal generic characters, varies particularly in B. mon- 

 tanus, the series of which is the largest, from a sharp, almost tooth- 

 like process to an entire absence of it. 



In order that the resemblance in form of the difierent varieties may 

 be most readily seen, I have placed all the measurements in the same 

 table, instead of giving those of each variety, with the description of 

 its plumage. The specimens in the collection of the Smithsonian In- 

 stitution are designated by the number of their labels ; those from the 

 Academy by the letter A, and a few from my own collection by the 

 letter B. Though the sex, length, and extent are copied from the 

 labels when mentioned, much value should not be attached to them 

 as they are obviously wrong in some instances ; in no case is the 

 measurement of the length of the prepared skin given, as I do not 

 consider that there is any certainty of its approximating very nearly 

 to the real length. Apparently the length of the red-tailed Hawks 

 varies from 600 to 700 millimetres, and of the other species from 550 

 to 650 miUimetres. 



Buteo horealis, adult. Above, dark brown, with purple reflections ; 

 base of the feathers of occiput, hind-neck and forehead, white ; 

 feathers of the back, scapulars, and wing-coverts generally with 

 their margins lighter, as if faded, and sometimes ferruginous ; feathers 

 of hind-neck very dark, with their margins ferruginous ; sides of neck 

 and temporal regions the same, frequently appearing rufous, streaked 

 with brown. Top of head slaty brown, margins of the feathers 

 rufous ; lores white, the bristly ends of the feathers only being black. 

 Closed wing brown, with the ends of primaries dark brownish-black, 

 narrowly margined at the ends with whitish ; secondaries, tertiaries, 

 and centre of primaries obscurely barred with darker brown ; upper 

 tail coverts varying from white to rufous, and barred more or less 



