582 MR. R. B. SHARPE ON CERTAIN ACCIPITRES. [Nov. 3, 



In its dark coloration it approaches closely to a specimen of G. japo- 

 nica ; but this bird has not such a thoroughly blue tail. 



I add a few measurements of Kestrels, as those given in my ' Cata- 

 logue ' do not give an exact idea of the proportions of G. tinnunculus 

 and C. rupicola, the former appearing rather too small. 



a. C. tinnunculus. 



Total length. Wing. Tail. Tarsus. 



1. c? ad. Thuringia 12-5 9"6 67 1*4 



2. $ ad. Aboyne, N.B. .. 140 10-2 7"0 1-5 



3. (J ad. Belgium 14-0 10*0 7*5 1*5 



4. d ad. Nepal 14-0 103 7'3 1-6 



5. rf ad. Behar 15-0 104 7'3 1-6 



6. cJjuv. Bagdad 13-0 9-4 6*6 1-55 



/3. C. tinnunculus (dark race). 



1. $ ad. Aldenham, Herts. 14-0 9'8 7*0 1*55 



2. $ imm. Fokien, China. 14-5 102 67 1*6 



y. C. rupicola. 



a. $ ad. Cape of Good Hope. 127 97 6-0 1-45 



b. 2 ad. Angola 140 96 6-6 1-45 



c. dad. „ 12-0 9'2 6-2 1-45 



d. dad. „ 12-0 8-8 5-9 1'4 



e. dad. Cape of Good Hope. 125 10-1 6-2 1-55 



Kestrels, like other raptorial birds, are never very easy to measure ; 



and it is seldom that two people measure the same bird with exactly 



the same results. The dimensions of these birds, therefore, can only 



be taken in a very broad and general sense ; but supposing that in 



the above list we have an average series of specimens, the following 



result is obtained : — 



Male. Female. 



„ .. . f Wing 9-4-10-4. Wing 9-8-10-2. 



C. tinnunculus.... j Targ = s ^.g Targ ^ ^j.g 



. . fWing 8-8-10-1. Wing 9-6-9-7. 



C - rw P lcola 1 Tarsus 1-4-1-55. Tarsus 1'45. 



In the ' Catalogue ' (p. 428) a series of C. tinnunculus measured 

 as follows:— (6*) wing 9-3-10-2, tarsus 1-45-1-6; (?) wing 9-5, 

 tarsus 1-6 ; so that the full results of my measurements of this spe- 

 cies show that the wing of the male varies from 9'3 to 10 - 4 inches, 

 and its tarsus from 1*4 to 1*6 inch; while in the female the wing 

 varies from 9-5 to 10 f 2 inches, and the tarsus from 1*5 to T6 inch. 

 I discard the dimensions of C. rupicola given by me in the above- 

 mentioned volume, as some error has certainly crept in there. The 

 general average, therefore, of C. tinnunculus is larger than that of 

 C. rupicola, although both species vary immensely. On looking 

 over a series of both placed side by side, almost the only differences 

 are the darker and more chestnut tone of the rufous in C. rupicola 

 as compared with the paler and more vinous tint of C. tinnunculus, 



