STRUCTURE AND RELATIONSHIPS OF OPISTHOCELIAN DINOSAURS. 187 
expanded and was received by a distinct facet on the mesial angle of 
the ilium. 
In this codssification may be recognized the initial stage of the 
conversion of a presacral rib into an ilium-supporting element. As 
the head and tubercle united with the vertebra and became immov- 
able, and the distal.end came in contact with the ilium, ligamentary 
attachment would naturally result. Consequent stress upon the 
attached rib would lead to specialization in order to meet the new 
function laid upon it. 
The dorso-sacral rib bears evidence of just such a modification, 
which has been carried much farther (Fig. 1, Plate xivi). The 
diapophysis of the vertebra has been considerably reduced and over- 
laps the tubercular portion of the rib with which it has fused. The 
latter articulates distally with the crest of the illum. The head of 
the rib has fused completely with the parapophysis at the latero- 
superior angle of the centrum. The distal portion of this element of 
the rib probably forms the stout process which abuts the great peduncle 
of the ilium, although this cannot be positively determined from 
the specimen under consideration. But there can be no doubt that the 
dorso-sacral of this genus is a modified presacral vertebra. 
The position of the capitular attachment of the ribs on the lateral 
surface of the vertebral centra, as observed in the anterior dorsals, 
sacrals, and caudals, is evidently the primitive one. From that point 
the facets have been thrust upward by the dilation of the posterior 
thoracic and floating ribs. The position of the capitular attachment 
at the superior angle of the centrum in the dorso-sacral vertebra 
implies that the rib has either become fixed in its acquired function at 
a period in ancestral development, when the attachment of the last rib 
had reached this point in its upward progress, or that, having been 
elevated, it was again depressed in order to meet the stress of this new 
function. The amount of modification in the spine, zygapophyses, 
and centrum of this vertebra indicates that its union with the sacrum 
has taken place at a comparatively recent period. | 
CAUDAL VERTEBRZ. 
The series of caudal vertebre as represented by the Museum speci- 
men is not essentially different from that figured in Marsh’s restoration 
of ‘‘Brontosaurus.’’ (See Plate tu.) Dr. Osborn has insisted * that 
eight or ten of the anterior vertebrz were omitted in Marsh’s restora- 
tion, and has also estimated t that seven anterior vertebrz were miss- 
* Memoirs of the American Museum of Natural History, Vol. i, Part v, p. 213. 
+ Bull. Am. Mus. of Nat. Hist., Vol. x, p. 224. 
