178 MR. W. K. PARKER ON STEATORNIS CARIPENSIS. [Apr. 2, 



millira. transversely ; the arrest of the keel, behiud, is not seen in 

 Caprimulgus. 



The scapulae of Steatornis (Plate XVIII. fig. 8, sc.) are 50 

 millim. long, .5 millira. broad at their widest part, and 3 millim. in 

 the middle ; they are elegantly xiphoid, and have a sharp decurved 

 point. The coracoids (cr.) are also long ; their length is 38 millim., 

 the upper, or fifth, 8 millim.; at their largest (epicoracoid) angular 

 expansion they are 15 millim. across, and only 3*5 millim. in the 

 middle of the elegant and slightly sigmoid shaft ; the clavicular 

 process (root of precoracoid) is very small. 



The furcula (jFr.) is strong and U-shaped ; in a straight line from 

 the apex to their lower junction the rami measure 37 millim. ; their 

 average growth is 3 millim. ; it is very uniform. 



There is a small interclavicular knob (i.cL), 2 miUim. high, behind 

 the junction of the rami. 



The aborted "precoracoid" segment of cartilage has added very 

 little to the apex of each ramus. 



The curve of the rami (c/.) is great, and the roundness of the space 

 where they meet below is perfect ; it is a typical U-shaped furcula. 



This is one of the most common forms of the furcula, not only in 

 the great and varied group of the Coccygomorphae, but also in Birds 

 generally. Outside the Passerines, in the Arboreal groups, this 

 form is very common, e. g. in Trochilidae, Cypselidse, and in Po- 

 dargus, Eurystomus, and Bucerotidse ; whilst Rhamphastos, with its 

 divided rami, and the Picidae and most of the Alcedinidse, have Pas- 

 serine apices to their rami, that is they are dilated or bilobate. 



V. The Wing. 

 The extreme slenderness of the bones forming the wing oi Stea- 

 tornis (Plate XX. fig. I) is in great contrast with what we see in the 

 Bucerotidse, with their dilated and cellular bones. 



The relative length, however, of the three regions is very similar 

 iu both, the humerus and the manus being both very short, the cu- 

 bitus very long. 



Thus, although this bird, like the Swifts and Humming-birds, has 

 its wings twice as large as its legs, it is not a Macrochire ; its hand 

 is very small. 



The following measurements of the region in several long-winged 

 birds will make this plain ; the meaning of these differences will be 

 self-evident when the mode of flight of each type is considered : — 



Humerus. Ulna. Manus. 



millim. millim. millim. 



Steatornis caripensis .... 72 106 82 



Buceros rujicollis 102 147 85 



Caprimulgus europceus . . 35 45 50 



Cypselus apus 12 18 43 



Strix flammea 80 88 75 



Larus canus 92 102 102 



It will be seen at once that the intensest specialization has 

 occurred in the Swift, where the humerus is only two thirds the 



