251 E. L. Troxell — Entelodonts in Marsh Collection. 



above the surface, allowing the passage underneath of 

 water or of small animals. ' ' 



Without commenting on this idea of Professor Cope, 

 let us examine other hypotheses, perhaps equally fantas- 

 tic. When we compare the skulls with those of Hippo- 

 potamus, we wonder whether they might have had large 

 thick lips, somewhat prehensile, which the tubercles by 

 their wide extent aided in movement: to pull the lips 

 away from the long teeth before biting and to draw the 

 corners of the mouth forward in order to fold long reeds 

 and grasses into the teeth. 



There is a striking similarity between the anterior men- 

 tal tubercle in A. davits clavus and the protecting flange 

 for the canine in some of the ancient sabre-tooth cats and 

 in the Dinocerata, in that it forms a recess with a bony 

 projection beneath. In some of the species of AtcJke- 

 otlierium, and more especially in Pelonax, these guards 

 are very extensive. In Pelonax we do not know the 

 canines, but I imagine they were not so developed as to 

 require such a very long sheath for their protection. The 

 correlation between the wide-reaching tubercles and the 

 widespread points of the upper canines, especially where 

 the body of the ramus becomes smaller and the symphysis 

 shorter, is well illustrated in A. marshi (fig. 12), where 

 we find the points of the canines separated a distance 

 about equal to the width of the tubercles. 



The posterior mental processes were undoubtedly for 

 muscle attachment, and because they are generally curved 

 backward, we assume that the pull of the tendon was from 

 that direction or from above. Two possibilities present 

 themselves : first, from comparison with the dog (these 

 entelodonts resemble the carnivores in many respects), 

 we presume there was a muscle, homologous to the occipi- 

 to-mandibularis, having its origin on the paroccipitals, or, 

 in these animals, on the outer extremity of the temporal, 

 and its insertion on the ventral border of the ramus, 

 whose purpose it was to open the mouth or drop the jaws. 

 Secondly, there may have been a muscle corresponding 

 to the buccinator which in the dog is inserted low on the 

 ventral border of the ramus and has its origin on the 

 maxilla. In the Entelodontidae a muscle of similar nature 

 may have had its origin on the dependent process of the 

 jugal, its insertion on one or both of the mental processes, 

 and may have served the purpose, first, of a cheek muscle 



