G. GENERAL NATURAL HISTORY AND ZOOLOGY. 289 



remains were found portions of the jaws of an animal having 

 decidedly rei^tilian affinities, both jaws containing numerous 

 teeth, implanted in distinct sockets. These were small, com- 

 pressed, and pointed, and all similar in shape. There were 

 about twenty teeth in each ramus of the lower jaw, and 

 probably as many in the upper. The series extended over 

 the entire margin of the dentary bone, the front teeth being 

 near the extremity. 



A careful separation of all the parts of the matrix of this 

 specimen now satisfies Professor Marsh that the jaws really 

 belonged to the bird remains; and that we have here an in- 

 stance of the long-sought-for link between the bird and the 

 reptile, still more decided than in Archmopteryx, The dis- 

 tance between the two had been narroAved by successive 

 discoveries, and is now bridged over by this remarkable 

 fact. 



The greater part of the skeleton of this genus, according to 

 Professor Marsh, is generally bird-like in character, the lower 

 jaw being long and slender, and abruptly truncated just 

 behind the articulation for the quadrate bone. This extremi- 

 ty, and especially the articulation, is said to be very similar 

 in character to that of some recent aquatic birds. The skull 

 is of moderate size, and the eyes placed well forward. The 

 scapular arch and the bones of the wings and legs are strictly 

 ornithic ; and the sternum has a prominent keel and elonga- 

 ted grooves for the expanded coracoids. The metacarpals 

 are united, as in ordinary birds. The bones of the posterior 

 extremities resemble those of swimming birds. The verte- 

 brae, however, are all biconcave, the concavities at each end 

 of the centre being distinct and nearly alike. Although it 

 has not yet been possible to determine the fact, it is surmised 

 that the tail was elongated, as in Archceopteryx. Its size 

 was about equal to that of a pigeon. A supposed second 

 species bi Ichthyornis^ described as I. ceUi\ proved, on subse- 

 quent examination, to be generically distinct, and has been 

 termed Apatornis celer, and is made the type of the genus 

 Ap>atornis. Upon the two Professor Marsh has founded a 

 new sub-class, Odontornithes^ and the family Ichthyornidce. 



It is, however, not impossible that both genera are suf- 

 ficiently related Xo Archmopteryx to warrant their being like- 

 wise included in the sub-class Saurura?, although it is not 



N 



