26 NATURAL HISTORY OF BIRDS. 



tially as in living birds, as apparently was the sternum also. Of the clavicles only a 

 fragment of the upper end has yet been found. The wings clearly indicate very 

 strong power of flight. The humerus had an enormous radial crest, surpassing in 

 comparative size that of any living bird, and was placed in a plane nearly parallel with 

 the long axis of the head of the humerus, instead of considerably inclined, as in most 

 birds, thus strongly resembling, in these two points, the humerus in the pterodactyls, 

 the extinct flying reptiles. The carpal bones were two in number, and the metacar- 

 pals united as usual ; a noticeable feature of the second linger is that the thin lateral 

 expansion of the first phalanx ends in a prominent flattened, hook-like process beyond 

 the rest of the bone. 



The pelvic arch exhibits some interesting reptilian characters. The sacrum con- 

 sisted of about ten vertebra thoroughly anchylosed, as were also the pelvic bones. 

 Of these the ischium was expanded in the middle, extended further back than the 

 ilium, and was not united with the latter posteriorly, agreeing in that respect with 

 Hesperornis and a few modern reptilian birds. The legs and feet do not differ more 

 from those of modern birds than did the wings. The metatarsals are anchylosed 

 firmly and present no peculiar features. The phalanges, with the exception of one, 

 have not yet been found. 



That Ichthyornis was provided with feathers is proved beyond question by the 

 tubercles for the attachment of quills on the forearm. It will thus be seen that Ich- 



. 10. FIG. 11. FIG. 12. 



Vertebrae of Ichthyornis. 



thyornis, " the fish-bird," as it is fitly called from its fish-like vertebrae, was a remarkable 

 combination of very old and very modern characters, biconcave vertebrae and large head 

 with separate lower jaw and teeth, in connection with anchylosed metacarpals and 

 metatarsals. Referring to the accompanying cut (Fig. 9), which represents Marsh's 

 restoration of one of the species, for information concerning the general aspect of the 

 bird, we may remark, however, that the missing parts are supplied from a tern, a rather 

 specialized modern bird, and that consequently many features of the restoration are 

 unreliable, while one, at least, is manifestly incorrect. For we may safely assume that 

 Ichthyornis was holorhinal like Hesperornis, and not schizorhinal like a tern, as repre- 

 sented in the figure, and it seems rather strange that the head has been restored after 

 the fashion of the latter, when it is admitted that it resembles that of the former 

 "more nearly than that of any other known bird." We may, perhaps, also take excep- 

 tion to the restoration of the neck, as not in harmony with the disproportionate large 

 head. 



The gap between Ichthyornis and all other birds is very great, so it would be quite 

 unsafe to advance any opinion as to its genesis and relationships. All that we can say 

 at present is, that it sprung very early from the ancestral stock, preserving the primi- 

 tive character of the vertebrae and the skull long after other parts had reached an 

 advanced specialization, thus adding new evidence to the principle, " that an animal 

 may attain great development in one set of characters, and at the same time retain 

 other low features of the ancestral type." 



