INTEODUCTION. xxiu 



from the Creodonts to the true Zeuglodonts. In the earliest type, Protocetus, the 

 skull is already practically that of a Zeuglodont, the rostral portion being, in fact, even 

 more elongated than in some of the later forms ; at the same time, however, the 

 opening of the nares is situated nearer to the end of the snout. The teeth are most 

 remarkable; the incisors are not known, but the rest of the dentition is practically 

 that of a typical Creodont, none of the teeth having assumed the peculiar serrated 

 form characteristic of the later Zeuglodons. The canine is large, with a single 

 though grooved root ; the first premolar is much smaller, the second large and 

 two-rooted, the third and fourth have posterior cusps and are supported by three 

 roots ; the molars, which are small, are also three-rooted. In Prozeuglodon (p. 243) 

 from the later Birket-el-Qurun series the skull is likewise essentially Zeuglodont, 

 though the external nares have shifted a little further back. The canine is much 

 larger than the teeth before and behind it, and although the third and fourth 

 premolars and the molars have serrated crowns, as in the later Zeuglodons, they, or at 

 least the premolars, retain their inner buttress supported by a third root, so that in 

 tooth-structure this genus is exactly intermediate between Protocetus and Zeuglodon. 

 Fraas (29) has lately described still another annectant form, Eocetus, from about the 

 same horizon as Prozeuglodon, which attained a very large size ; the skull approaches 

 that of Zeuglodon in the position of the nares, but the teeth are Creodont-like and 

 possess inner (third) roots. Remains of Zeuglodon itself seem to have been found 

 in the same deposits as these intermediate types, but that genus continued to exist 

 in the Qasr-el-Sagha beds above after they had disappeared. 



The Birds are represented in these beds by one or two fragments only, but those of 

 considerable interest, because they show that most likely a true Eatite (JEremopezus) 

 existed in this district in the Upper Eocene period. This is of importance, because 

 it indicates that possibly some at least of the main subdivisions of the Eatitae may 

 have had a common ancestry in the Ethiopian region. If so, this would account for 

 the likeness between the ^Epyornithes and Struthiones referred to by Burckhardt, and 

 also suggests the possibility of the relationship between these groups and the South- 

 American Itheas. It is, of course, just possible that this genus is merely another 

 instance of the results of retrogressive modification leading to loss of flight and 

 increase in size in some group of Carinate Birds, such as has occurred in the case of 

 the Gastornithes and Stereornithes ; in any case, much more material is required 

 before any final conclusion can be reached as to the precise position of this genus. 

 The Eeptiles are represented by members of the orders Crocodilia, Chelonia, and 



d2 



