nA Shvfelvt, Osteology of Harris's Coymorant. ^^^' 



Oct. 



fusion, while the haemal spine of the 21st vertebra — also a long, 

 stout one — plays its part in forming the contour below of this 

 united bone (fig. 16, Plate XVII.) 



Between the vertebral element formed by the fused 20th and 

 2ist vertebrae there are two others in the dorsal region of the 

 spine and the pelvis. These are the 22nd and 23rd vertebra; of 

 the column, and they are entirely free. Their ribs meet the 

 sternum through elongate, stout costal ribs, as do the first pair 

 of ribs coming from beneath the ilia of the pelvis and the pair 

 on the 2ist vertebra. All of these ribs are large, highly developed 

 in every way, and support big epipleural appendages. The last 

 pair of the series, however, or the second pair which articulate 

 with the pelvic section of the column beneath the ilia, do not 

 possess " uncinate processes," nor do their flattened and very long 

 haemapophyses reach the costal border of the sternum on either 

 side. There is also a pair of " floating " costal ribs, each of con- 

 siderable length, and they articulate with the posterior border, 

 on either side, of the just-mentioned pair which precedes them. 



This arrangement of the rihs also obtains in P. urile, but all 

 the bones are slenderer and rather longer, and, all to the last pair, 

 their unciform processes are very large and elongate, while those 

 on the leading free pair of ribs of the series are conspicuously 

 expanded. As in the case of Nannopterum, the free, lower ends 

 of the ribs on the 19th and 20th vertebrae are " knobbed," while 

 the posterior borders of the third, fourth, and to some extent the 

 fifth pairs of ribs are sharpened, the lower part of which edge in 

 the case of the fourth pair being perceptibly extended. In Harris's 

 Cormorant these borders of the ribs are rounded. 



In P. vigua, P. aitritns, P. carbo, and doubtless other species, 

 this arrangement is essentially the same as we find it in Harris's 

 Cormorant, while slight differences may obtain in other species, 

 as, for instance, in P. punctatus the last pair of ribs — that is, the 

 second pair coming from beneath the ilia — have well-developed 

 epipleural appendages upon them, and this is the only Cormorant 

 I have met with that does have them on that pair of the series. 

 But then Phalacrocorax punctatiis of New Zealand presents a 

 number of other peculiarities in its skeleton, which will be touched 

 upon further on in this paper. 



Returning to the cervical vertebra of Nannopterum, we are to 

 observe, in the case of the atlas, that its neural arch is a flat, 

 smooth platelet of bone, broad and deep and nearly square in 

 outhne. Its postero-extemal angles are produced backwards and 

 outwards as conspicuous apophyses, while the neural canal is large 

 and transversely elliptical. The cup for the occipital condyle has 

 a circular perforation at its base, and the haemal spine is large, 

 somewhat inclined backward, has sharpened borders and minute 

 lateral processes above. These characters, with slight variations, 

 are repeated in the atlases of other Cormorants, though in P. 

 auritus the lateral apophyses on the haemal spine are unusually 

 well-produced. 



