24 NATURAL HISTORY OF BIRDS. 



necessary in order to obtain an intelligent idea of the state of that science, and of the 

 class it treats of, a full account of these ancestors of the feathered tribes has been 

 deemed desirable, and, as Professor Marsh's work is the only source of information, 

 the following statements are given as nearly in his own words as possible. 



The geological horizon of the known Odontornithes is in the middle cretaceous, 

 and corresponds to the strata named by Marsh the ' Pteranodon beds,' situated along 

 the eastern slope of the Rocky Mountains, and especially on the adjoining plains in 

 Kansas and Colorado. These beds consist mainly of a fine yellow chalk and calcare- 

 ous shale, both admirably adapted to preserve delicate specimens. The first bird fossil 

 discovered in this region was the lower end of the tibia of Hesperornis, found bv Pro- 

 fessor Marsh in December, 1870, near the Smoky Hill River, in western Kansas. In 

 June, 1871, he made the discovery of the skeleton which forms the type of II. regalis. 

 In the autumn, 1872, another skeleton of the same Avas found, and the type of the 

 genus Apatornis. The fossil birds procured in that region between 1870 and 1880, by 

 the different explorations, include remains of more than one hundred individuals of the 

 toothed birds. 



It was soon found that these toothed birds were of two different kinds, which, 

 although united under the common heading, Odontornithes, were more different than 

 almost any two living birds of the present day, and which had very little in common 



save the teeth. But even 



k h %||} ^jH ("H these were extremely dif- 



ferent, being placed, as 

 they were, in a continuous 

 groove in one group, while 

 in the other they were 

 implanted in individual 



FIG. 8. Quadrate bone of Ichthyornis. Sockets. The former Were 



therefore styled Odonto- 



holcaj (from the Greek odontoi, teeth, and holkos, a groove) while the others received 

 the name Odontotormae (odontoi, and tormos, a socket). The latter form the sub- 

 class here under consideration. 



The Odontotormae, or birds with teeth in sockets, so far as now known, were all of 

 small size, and possessed powerful wings and very small legs and feet. Some of their 

 characters as, for instance, their vertebra?, biconcave or hollow both behind and in 

 fi-ont separate them widely from all birds recent and extinct. 



The remains of this group preserved are more or less pneumatic, and this fact, in 

 connection with their small size, is perhaps the main reason why so few have been dis- 

 covered. As might naturally be expected, the hollow bones of flying birds, being- 

 filled with air, enable the carcass to float upon the water much longer than it other- 

 wise would, and it is thus liable to be destroyed by fishes or other animals. Hence, 

 the chances of complete skeletons being buried entire are greatly diminished. The 

 plains east of the Rocky Mountains have yielded remains of not less than seventy- 

 seven different individuals of Odontotormae, belonging to two well-marked "genera," 

 Ichthyornis and Apatornis, the former represented by several species (some of which 

 Avere formerly referred to the genus Graculavus), and the latter by only one. These 

 were all small birds, scarcely larger than a pigeon. In their powerful Avings and small 

 legs and feet, they remind one of the terns, and, according to present evidence, they 

 Avere aquatic birds, of similar life and habits. 



