210 D)'. C. W. Andreics — Prozeucjlodon atrox. 



the case with at least the first and second molars. The molars are 

 crowded together at the back of the jaw and seem already to be 

 undergoing the degeneration which in later forms leads to the loss of 

 at least some of them. The axis vertebra is very similar to that of 

 a carnivore, and the centra of the other cervicals are not shortened up 

 as in the later Zeugiodonts. Altogether Protocetus may be regarded 

 as a Creodont that has become adapted to an aquatic life. 



From beds of rather later age in the Fayum Mr. Beadnell collected 

 remains of another Zeuglodon, which has been described under the 

 name Prozeuglodon atrox. Plate IX is a photograph of a model of the 

 restored skull and mandible of this species, and gives a very good 

 idea of its chief peculiarities. The model has been skilfully con- 

 structed by Mr. F. 0. Earlow, mainly from a cast of the skull and 

 mandible which are the types of the species, but some details were 

 added from a less complete but comparatively undistorted specimen. 

 The skull on the whole is quite Zeuglodon-like, but the snout is a 

 little shorter than in Protocetus, and the nostrils a little farther back, 

 though not so far as in Zeuglodon. There is a strong high sagittal 

 crest, the tympanic bullae are well developed, and the zygomatic arch 

 is somewhat stouter than in later forms. The dentition appears to be 

 exactly intermediate between that of Protocetus from the beds below 

 and that of Zeuglodon osiris from those above. The dental formula 

 seems to have been i. -f, c. t. pm- f) ™- i" (?)• The incisors are conical, 

 sharply pointed teeth, separated from one another by moderate 

 intervals. The canine is larger than the incisors in front and the 

 premolar behindr, probably a relic of the Creodont condition. The 

 first premolar is a single-rooted tooth, and in one specimen it can be 

 seen that it must have been preceded by a milk-tooth. The other 

 premolars have tiie compressed serrated crowns characteristic of the 

 group. The second upper premolar has two roots, while the third 

 and fourth have a large third (inner) root as in Protocetus and 

 the Creodonts. The molars were probably two in number : it is 

 not clear whether their posterior root was single or imperfectly 

 divided into two. Another Creodont character in the upper jaw is 

 seen in the presence of deep fossae on the inner side of the premolars 

 for the reception of the points of the lower teeth. None of the lower 

 teeth have more than two roots. 



In beds of about the same age as those in which Pro%euglodon occurs 

 there are found two or three very large Zeugiodonts, one, Eocetus 

 Schweinfurthi, E. Fraas, being likewise in many respects intermediate 

 between Protocetus and Zeuglodon, while another, Zeuglodon isis, is, 

 so far as known, close to the true Zeugiodonts. The occurrence of 

 these large forms so early shows that specialisation in this particular 

 direction was very rapid. In the upper beds of the Middle Eocene 

 (Q,asr-el-Sagha series) of the Fayum only true Zeugiodonts are found, 

 differing in no important respect from the various species which at 

 that time had already spread over most of the seas, remains having 

 been found in Europe, America, and New Zealand. 



The great rapidity with which both the primitive whales and the 

 Sirenia seem to have become completely adapted to an aquatic life 

 is very suggestive. In both cases these groups appear to have arisen 



