PHYLOGENETIC 37 



these ancient birds were Ichthyornis and Hesperornis, from the 

 cretaceous shales of Kansas. 



On account of the presence of teeth in the jaws these have 

 been placed in a distinct " Sub-class " of birds— the " Odont- 

 ornithes," but it is open to question whether on such groui^jis 

 this is justified. 



Ichthyornis maj^perhaps be, and generally is, regarded as the 

 ancestral type of the present Steganopodes — the Gannets, Cor- 

 morants, Pelicans, Tropic and Frigate-birds. In all save the 

 teeth and the peculiar character of its vertebrae, which were 

 amphicoelous — reptilian characters — the skeleton of this bird 

 had acquired all the characteristic peculiarities of the Class Aves, 

 though we should expect, by the way, when the structure of 

 the palate is better known, to find that it presents even^ore 

 reptilian characters than are to be found in the living Struthious ^ 

 birds. 



Hesperornis, on the other hand, represents one of the most 

 highly specialised of all hirds, after the toothed jaws are taken 

 into consideration, having undergone the most radical changes 

 of structure in adaptation to an aquatic life (p. 385). 



These two birds, Ichthyornis and Hesperornis, though not 

 the only birds known from cretaceous rocks, represent the only 

 complete skeletons yet discovered, and it is significant that 

 they had already not only become adapted to different modes 

 of life, but that, in the case of Hesperornis, this adaptation had 

 attained a degree of specialisation exceeding that of any other 

 known bird, with the exception of the Moa, fossil or recent. 



The existence of these two very different types — the one a 

 bird of powerful flight the other not oply flightless but wing- 

 less, only a vestige^fW the upper arm remaining — points con- 

 clusively to a very extensive bird fauna at this remote period, 

 though of the land birds naturally but few would be preserved, 

 and these have not yet been discovered. 



Furthermore, when the whole of the available fossil material 

 comes to be examined, it is evident that the differentiation iflto 

 land and water birds took place in very remote times indeed, 

 dating from the Jurassic period, if not earlier. It is to the 

 earlier Jurassic formations then that we must look for traces of 

 the pro-avian types, for an insight into the beginnings of tj^e 

 evolution of the bird. And here, probably, if anywhere, we 



