Cory on the Birds of the West Indies. 207 



thigh-feathers broader and more rufous: under wing-coverts and 

 axillaries white, barred with pale rufous. Total length 17 inches, 

 culmen 1-65, wing 10-4, tail 8-6, tarsus 1-45. 



"Another specimen still quite young, agrees with the foregoing 

 in the coloration of the wings and tail, but has the edgings to the 

 feathers of the upper surface very much broader, and a broad white 

 tip to the tail; the sides of the face and coll;ir round the neck are 

 creamy white, without any brown spots; the under surface of the 

 body is also more free from spots, with here and there a feather 

 appeariag broadly barred with tawny rufous, indicative of the next 

 change in the plumage. 



"i1/(7/'«/-e;— Altogether different from the preceding stage. Above 

 leaden brown, the head more slaty, the sides of the face and chin 

 clear slaty blue; around the neck a rufous collar; quills brown, 

 with narrow apical margins of pale rufous or buffy white, the outer 

 secondaries rufous for nearly their whole extent, the under surface 

 of the wing greyish, creamy white near the base, all the quills 

 barred above and below with blackish brown ; tail ashy gi-ey, 

 crossed by two very broad bars of black, tippec^ with creamy white, 

 before which an indistinct subterminal line of ashy grey is visible, 

 some of the outer upper tail-coverts and base of tail slightly mottled 

 with whitish ; under surface of body tawny rufous, crossed with 

 b»oad bars of ochraceous buff, the under wing-coverts similarly 

 marked, the lower ones ochraceous buff, with greyish black cross- 

 bars. 



"The next change seems to be in the undersurface, where the 

 ochre-coloured become quite white, and whitish bars appear on the 

 grey throat. From this stage (to judge by our specimens) it 

 changes by a partial moult, and by a gradual change of feather at 

 the same time ; for the bars on the breast lose by degrees their 

 rufous tint and become grey, while the back also becomes slaty 

 grey instead of brown; the nuchal collar gradually disappears. 

 This gradual development seems lo be satisfactorily traced, with 

 the exception of the tail, which, instead of agreeing with that of 

 the rufous or "mature" stage, has four rather narrow black bars, 

 like the yoimg specimen first described. This can only be ac- 

 counted for by the fact that Hawks have really no fixed laws of 

 change in plumage, and that it is impossible for anyone to define 

 exactly the regular sequence of the variations. No two birds are ex- 

 actly alike; for one has the head more advanced another the tail, 

 vice versa. Thus the bird last noticed as donning iiis grey dress 

 is very far advanced as regards his body-plumage, but has not 

 moulted his tail, whereas those in the rufous dress are not so 

 forward in their body-plumage, but have already the tail of the 

 adult (one being in the act cf moulting"). 



"^f/«///c;««/t';— Slaty blue above and below; no trace of a nuchal 

 collar; under surface narrowly but irregularly barred with white, 



