84 THE CERATOPSIA. 



the adjacent faces of the two limbs of the L form a deep rounded sinus. These and the adjacent lateral surfaces are rough- 

 ened with grooves, some of them of large size, apparently for blood vessels. The convex side of the long limb is still rougher, 

 being transversely wrinkled and pierced by numerous pores. Its distal third is equally divided by a strong median groove. 



Bone No. 2 is composed of two elements, one of them entire, the other incomplete. The former consists of two triangular 

 plates, united by their longest borders so as to give a V on section and to inclose together a deep groove whose sides are elevated 

 at one end and gradually descend to the other. The line of junction is a narrow obtuse keel, and the external surface is fur- 

 rowed by grooves which are parallel to the shortest sides. This sheath bone incloses a slightly curved longitudinal element, 

 which extends freely from it at its long angle, as a rod with an oval section, and is nearly continuous with the keeled angle 

 of the embracing bone. In the other direction it becomes wider and deeper to the posterior border of the broken sheath bone. 

 Here it does not fill the sheath bone, but roofs over the inclosed space, which forms a conical axial cavity of the mass, which is 

 now filled with matrix. The surface away from the sheath bone is gently concave and is divided longitudinally by the base of 

 a septum or keel. The opposite surface of the free part of the median bone is equally divided by a longitudinal groove. 



Positive determination of these elements is at present impracticable, as they do not resemble the corresponding bones in any 

 animal known to me. No. 1 approximates in form the ethmoid of the gull (Larus), but appears, in part at least, to have been a 

 bone of the external surface. The long limb has nearly the appearance of those parts of the bird's skull which are inclosed in a 

 horny sheath. The inferior septum is not appropriate to that element. Its proper position at the front of the. basicranial axis 

 is less probable, because bone No. 2 is more appropriately placed there. If we then suppose No. 1 to be the septum narium and 

 adjacent part posterior to it, we are met with the anomalous recurved short limb of the bone, which thus becomes a horn-like 

 projection directed upward and forward at the base of the muzzle. This may be considered in connection with the rising pro- 

 jection of the supraoccipital bone, and with the fact that this short limb is entirely filled with moderately coarse cellular tissue. 

 As to bone No. 2, its sheath-like portion may be parasphenoid, and the axial part presphenoid or sphenoidal rostrum, or the 

 former may be the vomer and the latter the septum nasi or basitrabecular. 



From the preceding it is evident that the only comparisons which throw any light on the probable positions of these bones 

 are those made with cranial elements of birds. 



Bone No. 3 was found in contact with No. 2. It is flat and subparallelogrammic in shape. One side (the thickest) is 

 excavated by a regular arch, with smooth free border at right angles to the other surfaces. A part of the opposite side exhibits 

 a free narrow edge. All the other borders are sutural, generally partly squamosal, without serrature or roughness. This bone 

 is lateral, and the segment of a circle may be a portion of the orbit. There are several other bones belonging to this series, but 

 their description is postponed until their identification is practicable. No elements of the skeleton not cranial were found, 

 excepting a rib, a humerus, and a portion of the transverse border of the episternum. The latter resembles the corresponding 

 piece in the Monoclonius crassus Cope, and a similar fragment of large size was found with the remains of the Agathaumas 

 sylvestre in Wyoming. 



Owing to the fragmentary nature of this material, Cope was unable to determine with accu- 

 racy the homologies of the different parts preserved, and the species remained unnamed' until 

 after Marsh's publications on the Ceratopsia made known their true nature, when they were 

 referred to the above species, which was briefly described as follows in the American Naturalist of 

 August, 1889, page 716, published December 17, 1889: 



Monoclonius recurvicornis Cope, sp. nov. Dinosaurian Cope (Bull. U. S. Geol. and Geog. Surv. Terr., vol. 3, 1877, p. 588; 

 pi. 34, figs. 7 and 8). 



I excavated the bones of the skull of this species in Montana and described them as above, but suspecting that they might 

 belong to some of the species already known, I did not name them. The fortunate discovery by Professor Marsh enables me to 

 determine them. The supraorbital horns are robust, straight, and rather short. Their section is an antero-posterior oval at 

 base and at the middle rounded subquadrate. The nasal bones are wedge-shaped and much narrowed forward. They support 

 a coossified median septum below. Superior face rounded, very rugose. Some distance posterior to the apex they support a 

 ve^ robust horn, which is compressed and turned abruptly forward at the apex. Posterior face injured. Length of supra- 

 orbital horn 210 mm. ; long diameter at base, 115 mm. ; width of nasal bone at base of horn, 100 mm. ; diameter of nasal horn at 

 base (transverse), 95 mm. ; elevation (on curve) to broken apex, 115 mm. Between the supraorbital horns on each frontal bone 

 a low tuberosity. This was a colossal animal and of peculiar characters. The squamosal is narrower than in M. crassus and 

 had marginal tuberosities. 



In the original type a considerable portion of the basioccipital region is preserved, showing 

 the occipital, condyle formed by the coalesced basioccipital and exoccipitals which surround 

 and inclose the foramen magnum. The occipital condyle may best be described as pedunculate 

 and hemispherical in form. Its articular surface describes in fact an almost perfect hemisphere. 

 I have already stated that the occipital condyle in the Ceratopsia is formed by the united 

 basioccipital and exoccipitals. In the present specimen these elements are so firmly united 

 as to show no trace of the sutures; in several other instances, however, in younger individuals, 

 the sutures are open, thus affording conclusive proof as to the structure of the condyle, and 

 there can be no doubt that, as Marsh has stated, the basioccipital and exoccipitals unite to form 

 the occipital condyle, and that it is not formed by the basioccipital alone, as stated by Cope. 



