82 0. C. Marsh — Skull of Torosaurus. 



The parietal forms more than half of the upper surface of 

 the skull, and is the most characteristic element in its structure. 

 In the posterior part are two very large apertures, oval in out- 

 line, with their outer margin at one point formed by the 

 squamosal. The rest of the border is thin and somewhat 

 irregular, showing that the openings are true fontanelles. This 

 is still better seen in the second species represented in the same 

 Plate, fig. %,/''■ In the latter specimen, however, these vacui- 

 ties are entirely in the parietal, a thin strip of bone separating 

 them on either side from the squamosal. A second pair of 

 openings, much smaller, apparently the true supra-temporal 

 fossse, are shown in the type specimen. These are situated 

 mainly between the parietal and squamosal, directly behind the 

 bases of the large horn-cores. (Plate II, fig. 1, <?.) The same 

 apertures are represented in the genus Triceratops by oblique 

 openings, as in the skull shown on Plate III, fig. 4, where the 

 front border of each is formed by the post-frontal. 



Between these openings, in the type of Torosaurus, is a third 

 pair of apertures (Plate II, fig. 1, c'). These are quite small, 

 nearly circular in outline, and entirely in the parietal, although 

 probably connected originally with the supra-temporal fossae. 

 Another pair of still smaller foramina may be seen in the same 

 skull, close to the median line, and separated from each other 

 by the anterior projection of the parietal. A deep groove 

 leads forward to each of these foramina, along the suture be- 

 tween the parietal and post-frontal. The position and direction 

 of these perforations suggest that they may correspond to the 

 foramen seen in Triceratops, and shown on Plate III, fig. 4, x. 



The extreme lightness and great expanse of the posterior 

 crest in Torosaurus make it probable that it was encased in 

 the integuments of the head, and that no part of it was free. 

 The outer borders of both the parietal and the squamosals show 

 no marginal ossifications, as in the other known genera of the 

 group, but the presence of a large, separate, epijugal bone in 

 one specimen suggests that epoccipitals may yet be found. 



The open perforations in the parietal which have suggested 

 the name Torosaurus, readily separate this genus from all the 

 gigantic species hitherto known in the Ceratopsidm, but may 

 perhaps be found in some of the smaller and less specialized 

 forms, from lower horizons of the same formation. 



With the successive changes in the parietal there were corre- 

 sponding variations in the squamosals, and these bones also 

 will serve to distinguish the principal genera from each other. 

 In Plate III, the squamosals of four genera of this group are 

 shown, and the wide difference between them, when seen from 



