NO. 1606. OSTEOLOGY OF VAMPT0SAURU8—GILM0RE. 231 



consideration of these facts, there can be no question regarding the 

 association of these vertebra^. 



That the vertebra now under consideration was a dorsal is shown 

 by the sudden change of the capituhir facet from the anterior lateral 

 surface of the conti'iini on cervical nine, to a point well up on the 

 side of the arch beneath the transversely extended process or dia- 

 pophysis, and by the great development of the transverse process. 

 The capitular facet is also well developed as compared with the weak 

 facets of the cervicals anterior to it. This facet is slightly cupped 

 and subcircular in outline, l^eing somewhat elongated in the vertical 

 axis. 



The length of the centrum of No. 4282 is slightly less than in the 

 preceding vertebra. There is a pronounced cup on the posterior end 

 of the centrum, but anteriorly the end is less convex than in the 

 cervicals preceding, and, as Marsh has pointed out in his description 

 of the cervical region of Ceratosaurus, could only be inserted a short 

 distance into the adjoining cuj^. This distance is accurately marked 

 on the centrum by a narrow^ articular border, just posterior and ex- 

 ternal to the median flattened anterior face. ^Vllile there are no 

 lateral cavities in this centrum, the sides are deeply excavated, both 

 laterally and inferiorly. The inferior surface presents a narrow, 

 median, longitudinal ridge which widens at either end, more espe- 

 cially the anterior. The surface of this anterior expansion is rough- 

 ened with longitudinal stria?. 



The neural arch is high and incloses a large, circular neural canal. 

 The expanded pedicles of the arch are firmly attached to the centrum 

 by suture. Antero-posteriorly the arch, above its base, is considerably 

 .shorter than the centrum, the diapophyses rising from the sides of 

 the arch and extending upward and outward at an angle of 45°, as 

 .stout, subtriangular processes. The terminations of the diapophyses 

 are lacking in all available specimens. The anterior zygapophyses 

 look decidedly inward and slightly upward, and are elliptical in out- 

 line, with the greatest diameter antero-posteriorly. Connecting the 

 anterior zygapophyses at their inferior margins is a thick, rounded 

 shelf of bone which forms the covering of the anterior portion of the 

 neural canal. From the middle, and somewhat posterior to the an- 

 terior border of this shelf, a low spinous process is developed, which, 

 in No. 1877, Yale Museum, has a height of 11 mm. Superiorly its 

 termination is angularly roinided antero-posteriorly and slightly 

 thickened transversely. This spine is missing in all of the other 

 specimens studied. Posterior to the anterior zygapophyses and lat- 

 eral to the median spine are deep depressions in the top of the proc- 

 ess. The postzygapophyses are missing on the first dorsal of No. 

 4282, but are present on that of C. nanus, No. 2210. They extend far 

 back beyond the posterior end of the centrum, their articular faces 



