612 CYPSELUS MELBA. 



there being five on all, except tlic third, which has seven. 

 The claws are remarkably stout, well curved, much compress- 

 ed, concave beneath, very acute. 



The plumage is slightly glossed, and somewhat compact ; 

 the feathers ovato-oblong. The wings are extremely long and 

 narrow, extending nearly two inches beyond the tail ; the 

 second quill a quarter of an inch longer than the first ; the 

 secondaries extremely short, and only eight in number. The 

 tail is rather short, deeply emarginate, the lateral feathers 

 nearly an inch longer than those in the middle. 



The bill is black ; the iris hazel ; the toes dusky flesh-colour, 

 the claws dusky, with the tips pale. The general colour of 

 the plumage is greyish-brown, the quills of a darker tint, 

 their shafts greyish- black. There is a large patch of white on 

 the throat, and the breast and abdomen are of that colour ; the 

 rest of the lower parts, namely the sides and lower fore -part 

 of the neck, the sides of the body under the wings, the lower 

 wing-coverts, tarsal feathers, and lower tail-coverts, of the 

 same tint as the back. Most of the brown feathers, especially 

 those on the lower parts, are terminally edged with paler. 



Length to end of tail 9 inches, to end of wings lO^f ; ex- 

 tent of wings estimated at 21 ; wing from flexure 9j^ ; tail 

 Sjg ' bill along the ridge ~%, along the edge of lower mandi- 

 ble li% ; tarsus j\^ ; first toe j%, its claw ^*g ; second toe j'^ 

 its claw j% ; third toe j%, its claw |\ ; fourth toe *^g, its 

 claw j^-g. 



Female. — The female is similar to the male. 



Habits. — This species, which is easily distinguished from 

 the other, by its superior size, and the white so conspicious 

 on its lower parts, appears to rank no higher with us than as 

 a rare and accidental straggler, although it has been conjectur- 

 ed that it may breed on the southern coast of Ireland. An in- 

 dividual is recorded to have been shot in the Isle of Thanet, 

 in June 1820; another near Buckenham Church in Norfolk, 

 in September 1831 ; and a third was found dead near Saflron 

 Walden in Essex, in July 1838. A specimen was also ob- 



