THE EVIDENCE FROM MISSING LINKS. 161 



light. By the new discovery both Hesperornis and Ichthyornis were 

 found to possess teeth; the former (Fig. 81) having its curved teeth 

 (B) set in a common groove in the jaw-bones ; whilst Ichthyornis 

 (Fig. 82) makes a further advance towards perfection in dental 

 apparatus, in that its twenty or so teeth of each jaw were lodged 

 in distinct sockets. The importance of these facts as bearing on 

 true reptile-like characters in birds may, be readily imagined. No 

 living bird possesses even the semblance of teeth, if we except the 

 horny ridges of the Merganser's bill. Prior to Marsh's discoveries, 

 no fossil bird was known to have been provided with true teeth 

 although indeed, in certain bird-remains, described by Owen, from 

 the London clay (Eocene) of Sheppey, under the name of Odontopteryx 

 (Fig. 83), the jaws were provided with bony projections. These pro- 

 jections, however, are not true teeth which, as many readers may 

 know, do not resemble bones, either in development or structure, being 

 developed from the "gum" or lining membrane of the mouth, and not 

 from cartilage, as true bones usually are. Doubtless these bony projec- 

 tions aided Odontopteryx to catch its finny prey,..as the horny ridges 



FIG. 83. ODONTOPTERYX (RESTORED). 



of the Mergansers enable them to retain the.fishes they so dexterously 

 capture. One curious bird (Phytotoma), a South American Leaf- 

 cutter, certainly possesses a double row of bony projections on its 

 palate. But even this novel and unusual addition to the list of bird- 

 possessions bears but a faint resemblance to the bony teeth of Odont- 

 opteryx, as these latter in turn are an entirely different and relatively 

 modern feature of the bird- type, when compared with the true teeth 

 of their " American cousins " of the Western Chalk. 



The Ichthyornis, moreover, diminishes the distance betwixt birds 

 and reptiles in yet another fashion the joints of its spine (Fig. 82, B) 

 were concave at either end (<:), a conformation familiar to us in the joints 

 of the fish-backbone, utterly unknown in living birds, but common 

 enough in reptiles. This character alone, in the eyes of the naturalist, 

 becomes invested with an importance hardly to be over-estimated as 

 regards its reptilian relationships ; and in Hesperornis, also, certain 

 features in addition to those already noted show unmistakable marks 

 of affinity to the reptile type. The teeth of this latter bird, set, as 



M 



