608 PROF. H. G. SEELEY ON THE DINOSAURIA OF 
of the neural canal in the sacral region, where it becomes much wider 
than the entire diameter of a dorsal centrum, and the absence of 
any corresponding pectoral enlargement is strongly suggestive, not 
only that the hind limbs were relatively more developed than the 
fore limbs, but that progression was carried on by means of the 
hind limbs; and the slight increase in size of the centrums from the 
neck to the sacrum would support such a conclusion. 
Caudal Vertebre (P1 XXXIV. fig. 5).—The eight caudal vertebree, 
when placed together in close succession, measure ‘84 inches in length, 
and when so placed arrange themselves in a curve, which has “the 
convex side downward, while the dorsal and sacral regions form a 
curve in the opposite direction ; and this curve of the tail appears to 
be correlated with the elevation of the root of the tail well above 
the ground. 
Each centrum is just over an inch long, the last of the series being 
of the same absolute length as the first, though the vertical and trans- 
verse measurements have become greatly reduced. The first caudal 
of the series exactly corresponds in size with the posterior articular 
face of the last sacral vertebra. 
The anterior face of the second centrum is flattened, but slightly 
concave, is of subcircular outline, nearly 1,2, inch wide and nearly 
1,);inch deep. The posterior face (fig. 5 a)1 is ‘subtriangular, owing to 
the sides converging inferiorly and terminating in the oblique facet for 
the chevron bone, which is more than half an inch wide, and rounds 
upward into the intervertebral articular surface. ‘This surface is 145 
inch wide in its upper third, and nearly 1,2, deep to the base of the 
chevron articulation: it is much more concave than the anterior 
articulation, and the depression is similarly deepest just above a 
slight mamillate eminence in the centre of the intervertebral sur- 
face. The antero-posterior measurement of the centrum is | inch. 
The base of the neural arch Hes come away on the left side, leaving 
a large pit 2 inch long and ;4, inch wide. On the right side the 
pedicle has remained attached, ‘and shows that from this broad base 
a small compressed lamina ascended, directed inward, to form the 
arch over the neural canal. At the sides of the centrum are, on 
each side, an ovate facet more than half an inch long and of less 
depth, placed midway between the anterior and posterior articular 
faces of the vertebra. These facets gave attachment to the trans- 
yerse processes or caudal ribs; they have an elevated border, and 
the upper fourth is formed by the pedicle of the neural arch, while 
the remainder of the facet is on the upper part of the side of the 
centrum. In subsequent vertebree this facet decreases in size, de- 
scends a little in position, and is placed nearer to the posterior end 
of the centrum. ‘The sides of the centrum conmcnee inferiorly 
towards a flattened narrow base, which is about ;4, inch wide, and 
owes its existence and imperfect definition to the chevron bone at- 
tached to the base of the posterior articular surface. The ventral sur- 
face of the first centrum is considerably more convex from side to side. 
In the sixth caudal the width and depth of the centrum are about an 
inch ; ius antero-posterior extent of the narrow lamina of the neural 
arch is ,%, inch; the transverse process appears to be short, thick, and 
