10 
Specimen No. 8504 was not found, Mr. Sternberg informs me, in situ, 
but in the float, though all the bones were close together and it, therefore, 
seems quite probable, because of the apparent rarity of such vertebrae, 
that they pertain to a single individual. These are the only vertebrae of 
this particular kind that have been found by Geological Survey parties 
during their twenty-five years of exploration in the Upper Cretaceous 
deposits, and it should be added that in none of the many collections 
which have passed through my hands for identification for the United 
States Geological Survey during the past twenty years, do I recall having 
observed any of a similar nature. 
The smooth, dense, exterior surface of the bones, with coarsely can- 
cellated internal structure, and centra with deep, clearly-defined pleuro- 
central cavities and with gentle concave articular faces, appear to show the 
theropod affinities of these bones. 
The two dorsal vertebrae, though from widely separated geological 
formations, are almost identical in shape and structure, differing only in 
size. The smaller one, No. 8505, is from the Belly River formation, and 
the larger one, No. 8504, is from the Lance formation. 
The sides and ventral surfaces of these vertebrae are moderately concave 
longitudinally; articular ends gently concave; no trace of neurocentral 
suture, though the basal portions of the neuropophyses are present in both 
specimens. The most conspicuous and distinctive feature of these verte- 
brae is the deep lenticular pleurocentral fossae on the upper lateral surfaces 
of the centra. These extend inward and slightly downward nearly to the 
centre of the bone, though rapidly contracting in both dorso-ventral and 
longitudinal diameters. The neural canal is of good size and subcircular 
in outline. The larger dorsal, No. 8504, has a greatest length of 44 mm. 
and a greatest transverse diameter of 36 mm. 
In Figure 5 are shown lateral, ventral, and end views of the caudal 
vertebrae found associated with the dorsal, No. 8504, described above. 
These are regarded as belonging to the anterior half of the tail because of 
the presence of transverse processes which have their origin on the arch. 
These vertebrae are relatively short, with lateral cavities high up beneath 
the transverse processes. The articular ends of the centra are more 
decidely concave than in the dorsals, though they exhibit the same coarsely 
vesicular internal structure of the bone. The sides of the centra are very 
shallowly concave from end to end, and the narrowed ventral surface is 
traversed by a pronounced longitudinal groove, with a foramen-like pit 
toward the anterior end, which leads upward into the interior of the centrum. 
The neuropophyses, as in the dorsals, are all firmly coossified with the 
centra, all trace of their union being obliterated. Chevron facets are very 
indistinctly indicated, if present at all. These vertebra vary in length 
from 22 to 25 mm. The neural canal is circular in outline. 
The relatively small size of all of these vertebrae shows them to belong 
to one of the smaller carnivorous dinosaurs, but whether to any of the 
families now known appears very uncertain. The lack of pleurocentral 
cavities in either dorsal or caudal regions, and the presence of more elongate 
caudal vertebra in the Ornithomimidae would appear to exclude them from 
that family. Likewise their affinities do not appear to lie in the Coelurid®, 
