g8 Shufeldt, Osteology of Harris's Cormorant. [,,f "oct. 



in this skeleton, tor the reason that that vertebra always has 

 shorter transverse processes than the one next preceding it. In 

 No. 19,721 the caudal vertebrae include the 42nd to 48th, being 

 7 in ali, to which the pygostyle is to be added ; in 19,722 the 

 skeleton of the tail is probably in the skin of this specimen, which 

 was preserved, and so the count cannot be made. 



Finally, we have No. 19,628, in which, differing from all the rest, 

 although it has seven vertebras and the pygostyle in its tail, these 

 vertebrae are the 41st to the 47th inclusive. 



As to the number of vertebrce in the pygostyle, I should say that 

 there were about seven, although there can be no certainty about 

 it until they have been correctly counted in many young 

 Cormorants at different ages and proper stages of development. 

 There is no difficulty in making out the leading five vertebrae in 

 the pygostyle of a Cormorant, and the posterior pointed looks as 

 though it might contain two more. 



Harris's Cormorant has the caudal vertebrae and pygostyle of 

 great size — in fact, of comparatively massive proportions, with all 

 the main processes well developed. This is not the case with 

 many other species, where, although they are fairly well developed, 

 the caudal vertebrae rarely exceed six in number, and the added 

 pygostyle may be small in comparison. This is the case, to some 

 extent, in P. albiventris, while P. punctattts possesses seven caudals 

 and a good-sized pygostyle. Doubtless there is an individual 

 variation in the number throughout the family, while in general 

 terms it may be said that Nannopteruni usually has seven caudals 

 and a pygostyle, and most all other Comiorants but six and a 

 pygostyle. 



Good views of the pelvts of Nannopteruni harrisi are presented 

 in figs. 16, 19, and 20 of the plates accompanying this paper. This 

 part of the skeleton is also liable to vary in different individuals, 

 while in old birds of the species it is always a big, strong, and 

 massive bone. As a whole, it assumes the narrow, elongate form 

 seen among water-birds of the Cormorant class. 



Viewed fiom above, it will be observed that the pre-acetabular 

 part of the bone is very considerably less than the post-acetabular, 

 and that its spreading anterior portion is horizontally disposed, 

 the surface rapidly becoming sub-vertical as we pass backward 

 to the acetabulas. In front the surface is roughened and lined 

 for muscular insertion, and is often perforated here and there by 

 small vascular foramina. Either antero-lateral angle is rounded 

 off ; the anterior border exhibits a raised and definite emarginate 

 finish, with generally some small, blunt processes extending 

 directly forwards. We also find occasionally one or more con- 

 spicuous apophyses outstanding from the lateral borders of this 

 pre-acetabular portion of the pelvis, and one of these is shown 

 in fig. 20 of Plate XVHI., the left-hand side one having been 

 broken off. 



Anteriorly, there is a re-entering angle between the iha, far into 

 which projects the neural spine of the first " pelvic vertebra." 



