110 THE CERATOPSIA 



"After comparison with the above ceratopsians of Belly River, Two Medicine, Edmonton and 

 Lance age, this new genus and species appears to be distinct from all previously described genera. 

 It differs widely from Anchiceratops of the Edmonton with its ornate crest, yet it would appear to 

 belong to a geologic stage equivalent to the Edmonton. It differs widely in its elongate propor- 

 tions from Brachyceratops. It is naturally profoundly different from the older Belly River and 

 Two Medicine genera described by Lambe, Gilmore, and Brown, although it might be geologically 

 successive to the Eoceratops canadensis of Lambe in the Belly River stage." 



The "lateral jugal osseous horns" are not new structures as they are present in nearly all Cera- 

 topsia, and were named epijugals by Marsh. Here, they have reached a remarkable degree of 

 elongation which distinguishes them from those of most genera. They are, however, very marked 

 in some specimens of Protoceratops, notably No. 6433, No. 6439, and No. 6467 A.M.N.H., as 

 well as in the type of Anchicerato-ps ornatus, No. 5251 A.M.N.H., and to a less extent in Chasmo- 

 saurus kaiseni, No. 5401 A.M.N.H., and in Arrhinoceratops brachyops type, No. 5135 Ct. R.O.M. 

 This character is, therefore, one of degree and not of kind. This is true also of the nasal and supra- 

 orbital horns. 



Osborn's statement that the fontanelles are much more elongate than in Triceratops prorsus is 

 clearly a slip if he means the parietal fenestrae, for there are none in Triceratops prorsus or in any 

 other Triceratops species, a fact of which he is well aware. 



Pentaceratops seems to resemble Chasmosaurus on the one hand, and Torosaurus on the other. 

 With the former, it agrees in the rather short face and longer muzzle, in the long squamosals and 

 large fenestrae, and in the position of the nasal horn. It disagrees in the possession of long supra- 

 orbital horns which in Chasmosaurus are very short, with two known exceptions that of Chasmo- 

 saurus kaiseni type skull No. 5401 A.M.N.H., and the specimen No. 40 Univ. of Alberta. With 

 Torosaurus, the differences are all the results of a probable evolutionary trend. The elongate crest 

 and the short face are alike. The muzzle of Torosaurus was also probably somewhat elongated, but 

 as both of the known skulls are defective in this regard, an actual comparison cannot be made. The 

 fenestrae of Torosaurus are relatively smaller although still persistent, and the epoccipitals have 

 disappeared, while the epijugals are reduced. There is also a reduction of the nasal horn, which in 

 common with that of the contemporary Triceratops has shifted forward to lie over the anterior, 

 instead of the posterior, rim of the narial opening. The base of each supraorbital horn in Toro- 

 saurus extends further back in relation to the orbit, a tendency which is also seen in the Pentaceratops 

 specimen No. 1624 A.M.N.H. 



The only point of agreement with Eoceratops that I can find is the position of the supraorbital 

 horn immediately over the orbit. The shape of the horn is, however, very different. The squa- 

 mosal of Eoceratops is rather short, a feature which would seem to debar it from a Pentaceratops 

 ancestry; also the position of the nasal horn in the two genera does not agree. Eoceratops certainly 

 cannot displace Chasmosaurus, its contemporary, as a probable ancestor of Pentaceratops. 



With Anchiceratops also there is little agreement. The position of the supraorbital horn with 

 reference to the orbit is different, for in Anchiceratops the orbit lies under the anterior margin of the 

 horn. The squamosals, while long, are shorter than in Pentaceratops, and the fenestrae are much 

 smaller. But there is evidence of agreement in the peculiar prominences that characterize the 

 posterior dorsal surface of the Anchiceratops crest, seen also in Pentaceratops No. 1625 A.M.N.H. 

 The epijugals do show a rather striking similarity but one could hardly call those of Anchiceratops 

 jugal horns. 



Altogether, at least seven specimens of Pentaceratops are known, five of which were collected 

 by Charles H. Sternberg. Three of them, the type of P. stembergii, No. 6325, and another skull, 

 No. 1624, with the crest of a third, No. 1625, are in the American Museum, while the others, a 

 skeleton without the skull (except the lower jaw), and another skull, are in Upsala, Sweden, and 

 constitute the types of Pentaceratops fenestratus Wiman. Some material including a fine squamosal is 

 in the National Museum. 



