378 
BE. E. W. SUUEELDT’s MOEPHOLOOICAL 
Chcetura pelagica. 
3. The last caudal vertebra is the 35tli. 
4. Pelvis much as we find it in some 
Swallows ; leading sacral vertebra 
does not markedly project beyond 
ilia. 
5. Sternum untouched posteriorly ; 
possesses comparatively large costal 
processes ; small manubrium ; deep 
carina ; which latter and the body are 
always riddled with large vacuities. 
6. Os furcula a very broad U-shaped 
one, with lateral abutments at its 
heads, and with rudimentary hypo- 
cleidium ; the bone harmoniously 
proportioned for the rest of the 
skeleton. 
7. Coracoids much of the same form 
as we find them in the Swallows. 
8. Blade of scapula nearly straight. 
9. General aspect of the body skele- 
ton, aside from the unnotched ster- 
num and rather deep keel to it, 
like the Hirundinidce. 
Trochilus rufus. 
3. The last caudal vertebra is the 32nd. 
4. Pelvis peculiarly formed ; and two 
entire vertebrae project beyond the 
ilia (the 18th and 19th). 
5. Sternum unnotched posteriorly ; 
very small costal processes ; no 
manubrium ; comparatively a much 
deeper carina ; sternal body and 
keel never perforated by vacuities. 
6. Os furcula rather of a very broad 
Y-shaped variety, with small lateral 
abutments at its heads, and with 
rudimentary hypoeleidium, with 
the bone of hair-like dimensions as 
compared with others of the skele- 
ton. 
7. Coracoids very peculiar, as the 
tendinal canal is closed by bone, 
and the shaft perforated by a large 
foramen below it. Totally unlike 
the bone in the Cy-pseli. 
8. Blade of scapula bent at a marked 
angle at its posterior extremity. 
9. General aspect of the body skeleton 
has no exact counterpart among 
living birds, that the writer has as 
yet ever met with. 
Now a few words as to what the above table shows : first, it is 
evident that the spinal column of Swifts and Humming-birds is 
fundamentally different, both in the number and arrangement 
of the vertebrae. It should, however, be stated that upon going 
over a large number of specimens, I find that it is the 15th 
vertebra that first connects with the sternum by costal ribs, and 
not the 16th. as stated in my first contribution of 18S5. This 
gives the Trocliili 3 true dorsals, which is as small a number 
as any existing bird possesses ; I found the same number in a 
Californian Condor. Cypseli possess 5 true dorsal vertebrce. 
Some excellent characters, no doubt, are to be obtained from 
any bird’s sternum, but the more I look into it the more I am 
convinced that the facility with which we can say sternum 
2-notched, sternum unnotched, sternum 4-notclied (as the case 
may be) has almost proved a detriment to avian taxonomy, for, 
